{"id":51487569,"url":"https://github.com/romanyn36/agentic-ai-roadmap","last_synced_at":"2026-07-07T08:01:39.084Z","repository":{"id":365379452,"uuid":"1151601707","full_name":"romanyn36/agentic-ai-roadmap","owner":"romanyn36","description":"A comprehensive AI learning roadmap covering Python fundamentals, mathematics, machine learning, deep learning, LLMs, and agentic systems — focused on hands-on projects, practical tools, and real-world deployment.","archived":false,"fork":false,"pushed_at":"2026-02-15T18:39:48.000Z","size":43,"stargazers_count":102,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2026-06-17T05:32:41.122Z","etag":null,"topics":["agentic","agentic-ai-roadmap","ai","ai-roadmap","deep-learning","langchain","langgraph","lear","llm","machine-learning","mathematics","oopenai","roadmap","transformer"],"latest_commit_sha":null,"homepage":"","language":null,"has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/romanyn36.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null},"funding":{"github":"romanyn36","custom":["https://www.paypal.me/romanyn36","https://www.buymeacoffee.com/romanyn36"]}},"created_at":"2026-02-06T17:07:31.000Z","updated_at":"2026-06-16T15:36:26.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/romanyn36/agentic-ai-roadmap","commit_stats":null,"previous_names":["romanyn36/agentic-ai-roadmap"],"tags_count":null,"template":false,"template_full_name":null,"purl":"pkg:github/romanyn36/agentic-ai-roadmap","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/romanyn36%2Fagentic-ai-roadmap","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/romanyn36%2Fagentic-ai-roadmap/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/romanyn36%2Fagentic-ai-roadmap/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/romanyn36%2Fagentic-ai-roadmap/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/romanyn36","download_url":"https://codeload.github.com/romanyn36/agentic-ai-roadmap/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/romanyn36%2Fagentic-ai-roadmap/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":35219594,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-05-26T15:22:16.424Z","status":"online","status_checked_at":"2026-07-07T02:00:07.222Z","response_time":90,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["agentic","agentic-ai-roadmap","ai","ai-roadmap","deep-learning","langchain","langgraph","lear","llm","machine-learning","mathematics","oopenai","roadmap","transformer"],"created_at":"2026-07-07T08:01:37.180Z","updated_at":"2026-07-07T08:01:39.056Z","avatar_url":"https://github.com/romanyn36.png","language":null,"funding_links":["https://github.com/sponsors/romanyn36","https://www.paypal.me/romanyn36","https://www.buymeacoffee.com/romanyn36"],"categories":[],"sub_categories":[],"readme":"\u003cdiv align=\"center\"\u003e\n\n# 🚀 The Complete AI Engineer Roadmap 2026\n### *Agentic \u0026 Generative AI: From Zero to Deployment*\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"https://readme-typing-svg.demolab.com?font=Fira+Code\u0026weight=700\u0026size=22\u0026duration=3000\u0026pause=1000\u0026color=8B5CF6\u0026center=true\u0026vCenter=true\u0026width=600\u0026lines=Learn+AI+from+Scratch+to+Production;Build+Intelligent+Agentic+Systems;Deploy+AI+Agents+to+Production\" alt=\"Typing SVG\" /\u003e\n\u003c/p\u003e\n\n*A structured roadmap from absolute beginner fundamentals in Python and mathematics, through machine learning and deep learning, to building and deploying advanced agentic AI systems.*\n\n[![GitHub stars](https://img.shields.io/github/stars/romanyn36/agentic-ai-roadmap?style=for-the-badge\u0026logo=github\u0026logoColor=white\u0026labelColor=1a1b27\u0026color=8B5CF6)](https://github.com/romanyn36/agentic-ai-roadmap/stargazers)\n[![GitHub forks](https://img.shields.io/github/forks/romanyn36/agentic-ai-roadmap?style=for-the-badge\u0026logo=github\u0026logoColor=white\u0026labelColor=1a1b27\u0026color=06B6D4)](https://github.com/romanyn36/agentic-ai-roadmap/network/members)\n[![GitHub issues](https://img.shields.io/github/issues/romanyn36/agentic-ai-roadmap?style=for-the-badge\u0026logo=github\u0026logoColor=white\u0026labelColor=1a1b27\u0026color=10B981)](https://github.com/romanyn36/agentic-ai-roadmap/issues)\n[![License](https://img.shields.io/github/license/romanyn36/agentic-ai-roadmap?style=for-the-badge\u0026labelColor=1a1b27\u0026color=F59E0B)](LICENSE)\n\n\n\n\n\u003c/div\u003e\n\n---\n\n## 📋 About This Roadmap\n\nbefore we dive in,iam **Romani Nasrat**— AI Engineer @ Penny Software, been freelancing since 2023. [See full bio below](#-about-me)  \n\n\u003e **💡so this isn’t just a normal roadmap** — I’m sharing my actual **personal learning journey**, from college all the way to professional AI work. Every resource here is personally tested and verified based on what I studied and used to reach where I am today.\n\u003e You’ll find both theoretical and practical topics in mathematics, computer science, and AI, along with carefully chosen learning resources. Everything is organized step-by-step, so just keep learning and don’t worry.\n\u003e This roadmap represents a **4-year journey**, including **2 years of professional experience freelancing and full-time work**, so don’t rush. Take your time, okay? You won’t become great in two months — and that’s totally fine.\n\n\n\u003e let me say it in Arabic-Egyptian 😊\u003cbr/\u003e\n\u003cp style=\"color: green;\" dir=\"rtl\"\u003e\n\n\u003e **💡 الرود ماب دي مش مجرد رود ماب عادية** — أنا بشارك رحلتي الشخصية الحقيقية في التعلم، من أيام الكلية لحد الشغل في مجال الـ AI بشكل احترافي.\n\u003e كل مصدر هنا أنا جربته وذاكرته بنفسي، بناءً على اللي اتعلمته واستخدمته عشان أوصل للمستوى اللي أنا فيه دلوقتي.\n\u003e هتلاقي فيها توبيكس نظرية وعملية في الرياضيات وعلوم الحاسب والـ AI، مع مصادر تعليمية مختارة بعناية. كل حاجة منظمة خطوة بخطوة، \n\u003e الرود ماب دي بتمثل رحلة 4 سنين، منهم سنتين خبرة عملية فريلانسنج وشغل فل تايم، فما تستعجلش. خد وقتك، تمام؟ مش هتبقى محترف في شهرين ودي حاجة عادية تماماً.\n\u003e \u003c/p\u003e\n\n\u003e **🎯 Who is this for?**\n\u003e\n\u003e * Beginners starting from scratch with Python and math\n\u003e * Students transitioning into AI/ML careers\n\u003e * Developers who want to build AI models, deploy systems, or create intelligent AI agents\n\u003e * Anyone looking for a clear, structured path from fundamentals all the way to production-ready \n\n---\n\n\u003cdetails open\u003e\n\u003csummary\u003e\u003cb\u003e📚 Table of Contents - Click to expand/collapse\u003c/b\u003e\u003c/summary\u003e\n\n- [🚀 The Complete AI Engineer Roadmap 2026](#-the-complete-ai-engineer-roadmap-2026)\n    - [*Agentic \\\u0026 Generative AI: From Zero to Deployment*](#agentic--generative-ai-from-zero-to-deployment)\n  - [📋 About This Roadmap](#-about-this-roadmap)\n  - [Phase 0 – Foundations](#phase-0--foundations)\n    - [0.1 Python Programming](#01-python-programming)\n    - [0.2 Math for AI (The \"Intuition\" Level)](#02-math-for-ai-the-intuition-level)\n  - [Phase 1 – Introduction to AI](#phase-1--introduction-to-ai)\n    - [1.1 Introduction to AI and AI Agents](#11-introduction-to-ai-and-ai-agents)\n    - [1.2 Search Algorithms for Problem Solving and AI Planning](#12-search-algorithms-for-problem-solving-and-ai-planning)\n  - [Phase 2 – AI \\\u0026 Machine Learning \\\u0026 Deep Learning \\\u0026 Data Science fundamentals](#phase-2--ai--machine-learning--deep-learning--data-science-fundamentals)\n    - [2.1 Machine Learning](#21-machine-learning)\n    - [2.2 Neural Networks \\\u0026 Deep Learning](#22-neural-networks--deep-learning)\n    - [2.3 Natural Language Processing (NLP)](#23-natural-language-processing-nlp)\n    - [2.4 Introduction to Computer Vision](#24-introduction-to-computer-vision)\n    - [2.5 Data Preparation](#25-data-preparation)\n    - [2.6 Data Science \\\u0026 Analysis \\\u0026 Visualization](#26-data-science--analysis--visualization)\n  - [Phase 3 – Agentic AI Systems](#phase-3--agentic-ai-systems)\n    - [3.1 Foundations: Transformers \\\u0026 LLMs](#31-foundations-transformers--llms)\n    - [3.2 AI Agents: Concepts \\\u0026 Architectures](#32-ai-agents-concepts--architectures)\n    - [3.3 Building Agentic Systems (Tools \\\u0026 Infrastructure)](#33-building-agentic-systems-tools--infrastructure)\n    - [3.4 Monitoring, Evaluation \\\u0026 Cost](#34-monitoring-evaluation--cost)\n    - [3.5 Deployment Production Ready Systems](#35-deployment-production-ready-systems)\n  - [📂 Additional Resources](#-additional-resources)\n  - [🤝 Contributing](#-contributing)\n  - [⭐ Show Your Support](#-show-your-support)\n  - [👨‍💻 About Me](#-about-me)\n    - [Romani Nasrat](#romani-nasrat)\n  - [📬 Contact \\\u0026 Connect](#-contact--connect)\n\n\u003c/details\u003e\n\n---\n\n## Phase 0 – Foundations\n\u003cp style=\"color: red;\"\u003e\nNote: this is just a starting point. For me, it came after two years of college study and before I began learning AI, so I already had a background in computer science and programming. However, if you’re a complete beginner, you can start here as well by learning Python from scratch.\n\u003c/p\u003e\n\n### 0.1 Python Programming \nTo build robust AI agents, you need more than just \"scripts\". You need an engineering-first approach to Python.\nin this phase you will learn the fundamentals of Python programming, including data structures, OOP, and modern Python features that are essential for developing AI systems. This will give you the tools to write clean, efficient, and maintainable code for your AI agents.\n\n\u003cdetails open\u003e\n  \u003csummary\u003e🔧 \u003cb\u003e1. Foundations \u0026 Core Mastery\u003c/b\u003e\u003c/summary\u003e\n  \u003col\u003e\n    \u003cli\u003e\u003cb\u003eAbsolute Basics:\u003c/b\u003e Variables, Data Types (\u003ccode\u003eint\u003c/code\u003e, \u003ccode\u003efloat\u003c/code\u003e, \u003ccode\u003estr\u003c/code\u003e, \u003ccode\u003ebool\u003c/code\u003e), and Basic Operators.\u003c/li\u003e\n    \u003cli\u003e\u003cb\u003eFunctions:\u003c/b\u003e Functions, args, kwargs, parameters, return values, and scope.\u003c/li\u003e\n    \u003cli\u003e\u003cb\u003eSyntax \u0026 Logic:\u003c/b\u003e Loops, list comprehensions, and Python 3.10+ \u003ccode\u003ematch\u003c/code\u003e statements.\u003c/li\u003e\n    \u003cli\u003e\u003cb\u003eData Structures:\u003c/b\u003e Deep dive into \u003ccode\u003elist\u003c/code\u003e, \u003ccode\u003edict\u003c/code\u003e, \u003ccode\u003eset\u003c/code\u003e, \u003ccode\u003etuple\u003c/code\u003e, plus \u003ccode\u003ecollections.deque\u003c/code\u003e and \u003ccode\u003edataclasses\u003c/code\u003e.\u003c/li\u003e\n    \u003cli\u003e\u003cb\u003eFunctional Tools:\u003c/b\u003e \u003ccode\u003elambda\u003c/code\u003e, \u003ccode\u003emap\u003c/code\u003e, \u003ccode\u003efilter\u003c/code\u003e, and the \u003ccode\u003eitertools\u003c/code\u003e module.\u003c/li\u003e\n    \u003cli\u003e\u003cb\u003eException Handling:\u003c/b\u003e Try/Except/Finally, custom exceptions, and context-aware error handling.\u003c/li\u003e\n  \u003c/ol\u003e\n\u003c/details\u003e\n\n\u003cdetails\u003e\n  \u003csummary\u003e🏗️ \u003cb\u003e2. Advanced OOP \u0026 Design Patterns\u003c/b\u003e\u003c/summary\u003e\n  \u003col\u003e\n    \u003cli\u003e\u003cb\u003eOOP 4 pillars:\u003c/b\u003e Inheritance, Abstraction, Encapsulation, Polymorphism.\u003c/li\u003e\n    \u003cli\u003e\u003cb\u003eOOP Fundamentals:\u003c/b\u003e Mixins (critical for framework customization), and Composition.\u003c/li\u003e\n    \u003cli\u003e\u003cb\u003eMagic Methods:\u003c/b\u003e Understanding \u003ccode\u003e__init__\u003c/code\u003e, \u003ccode\u003e__call__\u003c/code\u003e, \u003ccode\u003e__repr__\u003c/code\u003e, and \u003ccode\u003e__getattr__\u003c/code\u003e.\u003c/li\u003e\n    \u003cli\u003e\u003cb\u003eDecorators:\u003c/b\u003e Building custom decorators loic (logging, timing, and agent retry logic,etc).\u003c/li\u003e\n    \u003cli\u003e\u003cb\u003eContext Managers:\u003c/b\u003e Resource management using \u003ccode\u003ewith\u003c/code\u003e statements and \u003ccode\u003econtextlib\u003c/code\u003e.\u003c/li\u003e\n  \u003c/ol\u003e\n\u003c/details\u003e\n\n\u003cdetails\u003e\n  \u003csummary\u003e⚡ \u003cb\u003e3. Modern Python (The \"Agentic\" Stack)\u003c/b\u003e\u003c/summary\u003e\n  \u003col\u003e\n    \u003cli\u003e\u003cb\u003eType Hinting:\u003c/b\u003e Using \u003ccode\u003etyping\u003c/code\u003e (Annotated, Optional, Union) for robust, self-documenting code. \u003ca href=\"https://docs.python.org/3/library/typing.html\"\u003ePython Typing Docs\u003c/a\u003e\u003c/li\u003e\n    \u003cli\u003e\u003cb\u003ePydantic V2:\u003c/b\u003e Data validation and structured outputs—the backbone of LLM communication.\u003c/li\u003e\n    \u003cli\u003e\u003cb\u003eAsyncio:\u003c/b\u003e \u003ccode\u003easync/await\u003c/code\u003e, event loops, and concurrent task execution for responsive agents.\u003c/li\u003e\n    \u003cli\u003e\u003cb\u003eTesting:\u003c/b\u003e Writing unit and integration tests with \u003ccode\u003epytest\u003c/code\u003e to ensure agent reliability.\u003c/li\u003e\n    \u003cli\u003e\u003cb\u003eThreading \u0026 Multiprocessing:\u003c/b\u003e Understanding the differences between threading and multiprocessing and when to use each.\u003c/li\u003e\n    \u003cli\u003e\u003cb\u003eGenerators \u0026 Iterators:\u003c/b\u003e Understanding how to use generators and iterators to process large datasets.\u003c/li\u003e\n  \u003c/ol\u003e\n\u003c/details\u003e\n\n\u003cdetails\u003e\n  \u003csummary\u003e🌐 \u003cb\u003e4. Connectivity \u0026 Data Extraction\u003c/b\u003e\u003c/summary\u003e\n  \u003col\u003e\n    \u003cli\u003e\u003cb\u003eHTTP Clients:\u003c/b\u003e \u003ccode\u003erequests\u003c/code\u003e and \u003ccode\u003ehttpx\u003c/code\u003e (for async) to interact with external tools and APIs.\u003c/li\u003e\n    \u003cli\u003e\u003cb\u003eData Serialization:\u003c/b\u003e Handling \u003ccode\u003eJSON\u003c/code\u003e, \u003ccode\u003eYAML\u003c/code\u003e, and \u003ccode\u003eMarkdown\u003c/code\u003e programmatically.\u003c/li\u003e\n    \u003cli\u003e\u003cb\u003eWeb Scraping:\u003c/b\u003e Utilizing \u003ccode\u003eBeautifulSoup4\u003c/code\u003e and \u003ccode\u003ePlaywright\u003c/code\u003e/\u003ccode\u003eSelenium\u003c/code\u003e for browser automation.\u003c/li\u003e\n  \u003c/ol\u003e\n\u003c/details\u003e\n\n\u003cdetails\u003e\n  \u003csummary\u003e📦 \u003cb\u003e5. Environment \u0026 Package Management\u003c/b\u003e\u003c/summary\u003e\n  \u003col\u003e\n    \u003cli\u003e\u003cb\u003eHigh-Performance Tooling:\u003c/b\u003e Using \u003ccode\u003euv\u003c/code\u003e for ultra-fast dependency management.\u003c/li\u003e\n    \u003cli\u003e\u003cb\u003eStandard Tooling:\u003c/b\u003e \u003ccode\u003epip\u003c/code\u003e, \u003ccode\u003evenv\u003c/code\u003e, and \u003ccode\u003epyproject.toml\u003c/code\u003e configurations.\u003c/li\u003e\n    \u003cli\u003e\u003cb\u003eObservability:\u003c/b\u003e Implementing structured \u003ccode\u003elogging\u003c/code\u003e to trace agent reasoning steps.\u003c/li\u003e\n  \u003c/ol\u003e\n\u003c/details\u003e\n\n\u003c!-- resources --\u003e\n\u003cdetails\u003e\n  \u003csummary\u003e📚 \u003cb\u003e6. Resources\u003c/b\u003e\u003c/summary\u003e\n  this books what i used to learn python\n  \u003col\u003e\n    \u003cli\u003e\u003cb\u003eBook:\u003c/b\u003e Python Crash Course\u003c/li\u003e\n    \u003cli\u003e\u003cb\u003eBook:\u003c/b\u003e Python All in one for dummies\u003c/li\u003e\n    \u003cli\u003e\u003cb\u003eDocs:\u003c/b\u003e [Python Docs](https://www.python.org/doc/)\u003c/li\u003e\n    \u003cli\u003e\u003cb\u003eDocs:\u003c/b\u003e [The Hitchhiker’s Guide to Python](http://docs.python-guide.org/en/latest/)\u003c/li\u003e\n    \u003cli\u003e\u003cb\u003eDocs:\u003c/b\u003e [Python 3 Quick Reference](https://pewscorner.github.io/programming/python3_quick_ref.html)\u003c/li\u003e\n    \nExtra resources for i studied during learning\nbasic python\n- [BFCAI intro to python sections](https://drive.google.com/drive/u/0/folders/1DSA4O8CCa7l_ZuUfQ7WfAnYD2MVZiYTX)\n- [Python 3 Quick Reference](https://pewscorner.github.io/programming/python3_quick_ref.html)\n- [Ned Batchelder: Facts and Myths about Python Names and Values](https://www.youtube.com/watch?v=_AEJHKGk9ns)\n- [Ned Batchelder: Loop Like a Native - While, For, Iterators, Generators](https://www.youtube.com/watch?v=EnSu9hHGq5o)\n- [Idioms and Anti-Idioms in Python](https://nedbatchelder.com/text/iter)\n- [Transforming Code into Beautiful, Idiomatic Python](https://www.youtube.com/watch?v=OSGv2VnC0go)\n\n- [Top 10 Mistakes That Python Programmers Make](https://www.toptal.com/developers/python/top-10-mistakes-that-python-programmers-make)\n- [PEP 8 - Style Guide for Python Code](https://peps.python.org/pep-0008/)\n- [Mark Smith: More Than You Ever Wanted to Know About Python Functions (EuroPython 2018)](https://www.youtube.com/watch?v=vIkpCOY-yGs)\n- [Variables and Scope](https://python-textbok.readthedocs.io/en/1.0/Variables_and_Scope.html)\n- [73 Examples to Help You Master Python's F-Strings](https://miguendes.me/73-examples-to-help-you-master-pythons-f-strings)\n- [Richard Haven: Python Decorators and Diversions](https://www.youtube.com/watch?v=bP2ougQ3LHA)\n- [Doug Hellmann: Regular Expressions Are Nothing to Fear](https://dhellmann.github.io/presentation-regexes-fear/#/)\n- [Regular Expressions (Regex) Tutorial: How to Match Any Pattern of Text](https://www.youtube.com/watch?v=sa-TUpSx1JA)\n- [Brandon Rhodes: All Your Ducks in a Row - Data Structures in the Std Lib and Beyond (PyCon 2014)](https://www.youtube.com/watch?v=fYlnfvKVDoM\u0026t=257s)\n- [The magic of python context managers](https://towardsdatascience.com/the-magic-of-python-context-managers-adb92ace1dd0/)\n- [are the tuble more effent in python ?](https://stackoverflow.com/questions/68630/are-tuples-more-efficient-than-lists-in-python)\n- [Filter a set for matching string permutations](https://stackoverflow.com/questions/44857962/filter-a-set-for-matching-string-permutations/44858212#44858212)\n- [The Absolute minimum every software developer absolutely, positively must know about unicode and character sets](https://www.joelonsoftware.com/2003/10/08/the-absolute-minimum-every-software-developer-absolutely-positively-must-know-about-unicode-and-character-sets-no-excuses/)\n\nintermediate python\n\n- [Python’s super() considered super!](https://rhettinger.wordpress.com/2011/05/26/super-considered-super/)\n- [Context Managers the Easy Way](https://www.youtube.com/watch?v=U2t2t_cpvoc)\n- [Python dataclasses will save you HOURS, also featuring attrs](https://www.youtube.com/watch?v=vBH6GRJ1REM)\n- [Floating Point Arithmetic: Issues and Limitations](https://docs.python.org/3.7/tutorial/floatingpoint.html)\n- [difference-between-getattr-and-getattribute](https://stackoverflow.com/questions/3278077/difference-between-getattr-and-getattribute/3278104#3278104)\n- [What is the difference between venv, pyvenv, pyenv, virtualenv, virtualenvwrapper, pipenv, etc?](https://stackoverflow.com/questions/41573587/what-is-the-difference-between-venv-pyvenv-pyenv-virtualenv-virtualenvwrappe/41573588#41573588)\n\n\nadvanced python\n\n- [Raymond Hettinger - Super considered super! - PyCon 2015](https://www.youtube.com/watch?v=EiOglTERPEo)\n- [Descriptor HowTo Guide](https://docs.python.org/2/howto/descriptor.html)\n- [Python Descriptors Demystified](https://nbviewer.org/urls/gist.github.com/ChrisBeaumont/5758381/raw/descriptor_writeup.ipynb)\n- [What are metaclasses in Python?](https://stackoverflow.com/questions/100003/what-are-metaclasses-in-python/6581949#6581949)\n- [Understanding Python metaclasses](https://blog.ionelmc.ro/2015/02/09/understanding-python-metaclasses/)\n- [Grok the GIL: How to write fast and thread-safe Python](https://opensource.com/article/17/4/grok-gil)\n\n\n  \u003c/ol\u003e\n\u003c/details\u003e\n\u003cbr/\u003e\n\n\n\u003cbr/\u003e\n\n### 0.2 Math for AI (The \"Intuition\" Level)\n*You don't need a PhD, but you need to understand the mechanics.*\n\n\u003cdetails open\u003e\n  \u003csummary\u003e📐 \u003cb\u003eMathematical Foundations\u003c/b\u003e\u003c/summary\u003e\nAI is fundamentally built on math and statistics, so having a strong understanding of these foundations is essential to becoming a good AI engineer.\nIf you're looking for a great place to learn, I highly recommend the Arabic YouTube channel of my college professor, \u003ca href=\"https://www.facebook.com/ahmed.hagag.71/\"\u003eDr. Ahmed Hagag\u003c/a\u003e. He’s excellent at explaining both academic and practical concepts in a simple way. Honestly, he’s one of those rare teachers who truly leaves a lasting impact on his students.\n  \u003cul\u003e\n  \n  - [Discrete Math, Dr Ahmed Hagag](https://youtube.com/playlist?list=PLxIvc-MGOs6gZlMVYOOEtUHJmfUquCjwz\u0026si=Kf0EwkN2LU9KfJDj)\n  - [Linear Algebra, Dr Ahmed Hagag ](https://youtube.com/playlist?list=PLxIvc-MGOs6iQXFnjF_STbhGdrZBphrv_\u0026si=qNlk9EU2S6hG4HRi)\n  - [Calculus | Math. (1) DrAhmed Hagag](https://www.youtube.com/playlist?list=PLxIvc-MGOs6gkSl_PPAVJpebKDLo-ijEC)\n  - [Differential Equations, Eng. Yousef Elbaroudy](https://youtube.com/playlist?list=PL0_haP3tFYsPqcOPezL4RIlWtn3B2wALN\u0026si=o50YO7jR9vMvdNqt)\n  - [Statistical Analysis](https://youtube.com/playlist?list=PLxIvc-MGOs6ilU3FPyJr3T-VkufZy2NGi\u0026si=rcDrbobmYZ0UsjU5)\n  - [Statistics \u0026 Probability Dr.Ahmed Hagag](https://youtube.com/playlist?list=PLxIvc-MGOs6gW9SgkmoxE5w9vQkID1_r-\u0026si=Ori6WNPTkBJiKC4D)\n  - Book: Discrete Mathematicsand Its Applications SEVENTH EDITION byKenneth H. Rosen (our college reference book)\n  \u003c/ul\u003e\n\u003c/details\u003e\n\n\n---\n\n\n## Phase 1 – Introduction to AI \n### 1.1 Introduction to AI and AI Agents\n\nthe goal of this phase is to give a high-level understanding of what AI is and why it matters, and to introduce the concept of AI agents. This will provide the necessary context and motivation for diving deeper into the technical aspects of AI in the subsequent phases.\n\n- What is AI?\n- What is an AI agent?\n- Thinking Humanly (The Cognitive Modelling approach)\n- Acting Humanly (The Turing Test approach)\n- Thinking Rationally (The Laws-of-Thought approach)\n- Acting Rationally (The Rational Agent approach)\n- Why is AI important?\n- applications of AI in various industries (healthcare, finance, transportation, etc.)\n### 1.2 Search Algorithms for Problem Solving and AI Planning\nthe goal of this phase is to introduce the concept of search algorithms and how they are used in AI for problem solving and planning. This will give you a solid foundation in the fundamental techniques that underlie many AI systems, including agentic AI.\n\n\u003cdetails open\u003e\n\u003csummary\u003e🔍 Search Strategies\u003c/summary\u003e\n- Search Problem Components (State, Actions, Transition Model, Goal Test, Path Cost)\n- Search Strategies:\n  - Uninformed (Blind) Search:\n    - Depth-First Search (DFS)\n    - Breadth-First Search (BFS)\n    - Uniform Cost Search (UCS)\n    - Iterative Deepening Search\n    - Bidirectional Search\n  - Informed (Heuristic) Search:\n    - A* Search\n    - Greedy Best-First Search\n    - Adversarial Search (Game Search)\n    - Minimax Algorithm\n    - Alpha-Beta Pruning\n\u003c/details\u003e\n\n\u003cdetails\u003e\n\u003csummary\u003e🧬 Genetic Algorithms GAs\u003c/summary\u003e\n\n- Definition\n- Main Motivation for Heuristic Techniques (Including GAs)\n- GA Overview and Principle\n- Stochastic Operators / Evolutionary Cycle Steps\n  - Initialization\n  - Parent Selection\n  - Recombination (Crossover)\n  - Mutation\n  - Survivor Selection (Replacement)\n  - Evaluation (Fitness Function)\n\u003c/details\u003e\n\n\u003cdetails\u003e\n\u003csummary\u003e🧠 Knowledge and Experts\u003c/summary\u003e\n\n- Knowledge\n- Experts\n- Rules as Knowledge Representation\n- Expert Systems\n- Components of a Rule-Based Expert System\n- Expert Systems vs. Conventional Programs\n\u003c/details\u003e\n\nResources:\n- [Search Algorithms, Eng. Yousef Elbaroudy](https://youtube.com/playlist?list=PL0_haP3tFYsNlXKVk5cD2jP-Ev8D6Lexv\u0026si=dXSSxo-m0O28wUbM)\n- [AI  Search Algorithms In Python ](https://youtube.com/playlist?list=PLPBnj6azlABatXqkOgE4-Suu2ucfax42F\u0026si=RDeCzYltDBK2PVZ6)\n- [CS361 Intro. to AI [in Arabic], Dr.Amr S. Ghoneim](https://youtube.com/playlist?list=PLsnvpvHuTUbAZr0n65TgytBK6bHdT33A7\u0026si=oKSzKN0SGjklNHsn)\n- Artificial Intelligence: A Modern Approach (3rd Edition) (our college reference book)\n- [BFCAI Artificial Intelligence](https://drive.google.com/drive/u/0/folders/1WiaxjiOW0vZvf932Fl4YNwR7q2jEiZ1m)\n\n---\n\n## Phase 2 – AI \u0026 Machine Learning \u0026 Deep Learning \u0026 Data Science fundamentals\nThis phase provides a clear overview of the core AI fields and branches, and introduces the key techniques and algorithms used across the discipline — including machine learning, deep learning, reinforcement learning, and data science. Together these topics form a solid foundation for understanding how AI agents work and for designing and building them.\n\n### 2.1 Machine Learning\n\nIn this section, you will learn the fundamentals of machine learning, including core concepts, common algorithms, model evaluation, and optimization techniques used to build reliable predictive models.\n\n\u003cdetails open\u003e\n  \u003csummary\u003e🧠 \u003cb\u003e1. Core Concepts\u003c/b\u003e\u003c/summary\u003e\n  \u003cul\u003e\n    \u003cli\u003e\u003cb\u003eFoundations:\u003c/b\u003e Supervised vs Unsupervised learning, train/validation/test split, bias–variance tradeoff, overfitting vs underfitting.\u003c/li\u003e\n    \u003cli\u003e\u003cb\u003eSupervised Learning:\u003c/b\u003e Linear \u0026 Logistic Regression, KNN, SVM, Naive Bayes, Decision Trees, Random Forests, XGBoost, LDA/QDA.\u003c/li\u003e\n    \u003cli\u003e\u003cb\u003eUnsupervised Learning:\u003c/b\u003e K-Means, DBSCAN, Hierarchical Clustering, PCA, ICA, Anomaly Detection.\u003c/li\u003e\n    \u003cli\u003e\u003cb\u003eModel Evaluation:\u003c/b\u003e Accuracy, Precision/Recall, F1-score, ROC-AUC, MSE/MAE, Confusion Matrix, Cross-Validation.\u003c/li\u003e\n    \u003cli\u003e\u003cb\u003eModel Improvement:\u003c/b\u003e Feature engineering, regularization, hyperparameter tuning (Grid/Random Search), handling imbalanced data.\u003c/li\u003e\n    \u003cli\u003e\u003cb\u003eEnsemble Methods:\u003c/b\u003e Bagging, Boosting, Stacking.\u003c/li\u003e\n    \u003cli\u003e\u003cb\u003eReinforcement Learning:\u003c/b\u003e Q-Learning, DQN, Policy Gradients, Actor-Critic.\u003c/li\u003e\n    \u003cli\u003e\u003cb\u003eTools \u0026 Libraries:\u003c/b\u003e Python, NumPy, Pandas, Scikit-learn, Matplotlib/Seaborn(visualization), Jupyter.\u003c/li\u003e\n  \u003c/ul\u003e\n\u003c/details\u003e\n\n\u003cdetails open\u003e\n  \u003csummary\u003e📚 \u003cb\u003e2. Resources\u003c/b\u003e\u003c/summary\u003e\n\n\u003cp style=\"color: green;\"\u003e\nThese are the resources I personally used to learn Machine Learning, and I highly recommend them\u003cbr\u003e\nI was kinda a clever student 😂 — I mostly studied from my college lectures and followed the course sections step-by-step, then added Andrew Ng’s specialization + random helpful articles and videos.\u003cbr\u003e\nSo there isn’t one fixed source… learning ML is more like exploring from multiple places.\n\u003c/p\u003e\n\u003cul\u003e\n\n- \u003ca href=\"https://www.coursera.org/specializations/machine-learning-introduction\"\u003eMachine Learning Specialization (3 courses) — Andrew Ng (Coursera)\u003c/a\u003e\n- \u003ca href=\"https://www.kaggle.com/romanyn36\"\u003eKaggle Profile (ML projects and notebooks)\u003c/a\u003e\n- \u003ca href=\"https://developers.google.com/machine-learning/clustering/clustering-algorithms\"\u003eClustering Algorithms — Google Developers\u003c/a\u003e\n- \u003ca href=\"https://www.youtube.com/playlist?list=PLPBnj6azlABapMXzdpFXBScfZerZygcrz\"\u003eAdvanced Machine Learning, Dr Ahmed Yousry (in Arabic)\u003c/a\u003e\n- \u003ca href=\"https://youtu.be/LJjorwMNtsc?si=Ipb5jNBqcOJD4PTw\"\u003eWhat is Machine Learning?\u003c/a\u003e\n- \u003ca href=\"https://youtu.be/wL4mOvoq2Rg?si=rvKGNwWMKlp4DQLj\"\u003eRandom Forests Explained\u003c/a\u003e\n- \u003ca href=\"https://youtu.be/6JQK03IGIvU?si=2nl61MMEtYBRrmUa\"\u003eK-Means Clustering\u003c/a\u003e\n- \u003ca href=\"https://youtu.be/pzC-yTtoYPc?si=-tSJ12-j1uOCQ7eD\"\u003eLogistic Regression\u003c/a\u003e\n- \u003ca href=\"https://youtu.be/gnEaSBmg_-M?si=Cz423_Ip7eZar9As\"\u003eHow ML Algorithms Work\u003c/a\u003e\n- \u003ca href= \"https://drive.google.com/drive/folders/1QXHiED_LkxZxR7n8A1vX3P17IAWaYDRe?fbclid=IwZXh0bgNhZW0CMTAAYnJpZBExQTJDV202SHQ2NXNXWlNFdXNydGMGYXBwX2lkEDIyMjAzOTE3ODgyMDA4OTIAAR4RH4fuO2F58XXdbQi9_fAAf4pjRspV0hrPCCGvTEnPAjSfVXlzI-1aZidhTA_aem_JI08bQN575vdJFjWYP4woQ\" \u003eAI/Ml Martials (all topics vodeos with source code examples)\u003c/a\u003e\n-  Book: Machine Learning for dummies\n-  Book: \u003ca href=\"https://home-wordpress.deeplearning.ai/wp-content/uploads/2022/03/andrew-ng-machine-learning-yearning.pdf\"\u003eMachine Learning Yearning (Technical Strategy for AI Engineers, In the Era of Deep Learning) by Andrew Ng\u003c/a\u003e\n-  Book: Machine Learning with R second edition by Brett Lantz ( our college reference book)\n- \u003ca href=\"https://www.deeplearning.ai/resources/#course-slides\"\u003eDeepLearning.AI Al courses slides and resources\u003c/a\u003e\n\u003c/ul\u003e\n\u003c/details\u003e\n\n\n### 2.2 Neural Networks \u0026 Deep Learning\n\nIn this section, you will learn how neural networks work from the ground up, then move to advanced deep learning techniques used in real-world AI systems.\n\n\u003cdetails open\u003e\n  \u003csummary\u003e\u003cb\u003e🔹 Neural Networks — Basics\u003c/b\u003e\u003c/summary\u003e\n\nLearn the core building blocks of deep learning:\n- what is a neural network and how it works\n- Perceptron, neurons, layers (input / hidden / output)\n- Activation functions (ReLU, Sigmoid, Tanh, Softmax)\n- Forward \u0026 backward propagation\n- Loss functions\n- Gradient descent\n- Backpropagation fundamentals\n- Overfitting vs underfitting basics\nGoal: Understand how a neural network learns mathematically before using frameworks.\n\u003c/details\u003e\n\n\u003cdetails\u003e\n  \u003csummary\u003e\u003cb\u003e🚀 Deep Learning — Advanced \u0026 Practical\u003c/b\u003e\u003c/summary\u003e\n \nApply neural networks to real-world problems and large-scale models:\n\n**Training \u0026 Optimization**\n- Epochs, batch size, iterations\n- Gradient descent variants (SGD, Momentum, Nesterov)\n- Optimizers: SGD, Adam, RMSProp, AdaGrad\n- Regularization: L1/L2, Dropout\n- Batch normalization, Layer normalization\n- Learning rate schedules, Early stopping\n\n**Architectures \u0026 Models**\n- **Computer Vision:** CNNs, ResNet, EfficientNet, VGG, MobileNet\n- **Sequential / Time Series:** RNNs, LSTM, GRU, BiLSTM\n- **Attention \u0026 Transformers:** Transformers, BERT, GPT, T5\n- **Generative Models:** Autoencoders, Variational Autoencoders (VAE), GANs, Diffusion Models\n- **Graph Neural Networks (GNNs)**\n\n**Practical Skills**\n- Model building, training, evaluation, debugging\n- Model saving/loading (checkpoints, serialization)\n- Fine-tuning pretrained models\n- Transfer learning\n- Multi-GPU / TPU training\n- Deployment basics (ONNX, TorchScript)\n\n**Frameworks \u0026 Tools**\n- TensorFlow / Keras\n- PyTorch\n- Hugging Face Transformers\n- CUDA \u0026 GPU acceleration\n\n\u003c/details\u003e\n\n\u003cdetails\u003e\n\u003csummary\u003e\u003cb\u003e📚 Resources\u003c/b\u003e\u003c/summary\u003e\n\n- [Neural Networks and Deep Learning By Andrew Ng](https://www.coursera.org/learn/neural-networks-deep-learning?specialization=deep-learning)\n- [Artificial Neural Network Ar](https://youtube.com/playlist?list=PLPBnj6azlABZt1fD0B64lSkG1QktaF6kj\u0026si=nPWbXAAyN3d6PMtt)\n- [Deep Neural Network using Keras Ar](https://youtube.com/playlist?list=PLPBnj6azlABaYCtajh-TOpZ4UKcvWYQCj\u0026si=aaEfTZR780Ne4k0z)\n- [what is a neural network?, Dr Ahmed Yousri Ar](https://youtu.be/XIM3mCzLPcE?si=bUySaWpqjlQuqYNM)\n- [geeksforgeeks](https://www.geeksforgeeks.org/deep-learning/neural-networks-a-beginners-guide/)\n- [Deep Learning , Computer Vision, NLP Hessham Assem](https://youtube.com/playlist?list=PL6-3IRz2XF5X-lzMZdmkvGAx1a3kIm7_I\u0026si=j-3Y1ZLLpCt6D9RS)\n- [AI/Ml Martials (all topics vodeos with source code examples](https://drive.google.com/drive/folders/1QXHiED_LkxZxR7n8A1vX3P17IAWaYDRe?fbclid=IwZXh0bgNhZW0CMTAAYnJpZBExQTJDV202SHQ2NXNXWlNFdXNydGMGYXBwX2lkEDIyMjAzOTE3ODgyMDA4OTIAAR4RH4fuO2F58XXdbQi9_fAAf4pjRspV0hrPCCGvTEnPAjSfVXlzI-1aZidhTA_aem_JI08bQN575vdJFjWYP4woQ)\n- [Deep Learning specializations Deeplearning Ai](https://www.coursera.org/specializations/deep-learning#courses)\n- \u003ca href=\"https://www.deeplearning.ai/resources/#course-slides\"\u003eDeepLearning.AI Al courses slides and resources\u003c/a\u003e\n\u003c/details\u003e\n\n### 2.3 Natural Language Processing (NLP)\n\nNatural Language Processing (NLP) is the field that focuses on enabling machines to understand, interpret, generate, and interact using human language. This section summarizes core concepts, tasks, models, tooling, evaluation, and practical considerations for building NLP systems.\n\n\u003cdetails\u003e\n  \u003csummary\u003e\u003cb\u003e🔹 Core Concepts\u003c/b\u003e\u003c/summary\u003e\n\n  - **Linguistic Levels:** Morphology, Syntax (POS, parsing), Semantics, Pragmatics, Discourse.\n  - **Representation:** Tokens, subwords (BPE/WordPiece/SentencePiece), characters.\n  - **Classical Methods:** Bag-of-words, n-grams, TF-IDF.\n  - **Embeddings:** word2vec, GloVe, fastText, contextual embeddings (ELMo, BERT-style).\n  - **Text Cleaning:** lowercasing, normalization, stopword removal, stemming/lemmatization, de-duplication, cleaning noisy text.\n\u003c/details\u003e\n\n\u003cdetails\u003e\n  \u003csummary\u003e\u003cb\u003e🔧 NLP Tasks\u003c/b\u003e\u003c/summary\u003e\n\n  - **Text Classification:** Sentiment analysis, intent detection.\n  - **Sequence Tagging:** POS tagging, Named Entity Recognition (NER).\n  - **Parsing \u0026 Syntax:** Dependency \u0026 constituency parsing.\n  - **Coreference \u0026 Anaphora Resolution.**\n  - **Information Extraction:** Slot filling, relation extraction.\n  - **Question Answering (QA):** Extractive, abstractive, open-domain.\n  - **Machine Translation (MT).**\n  - **Summarization:** Extractive and abstractive.\n  - **Dialogue Systems \u0026 Conversational AI.**\n  - **Text Generation \u0026 Controlled Generation.**\n\u003c/details\u003e\n\n\u003cdetails\u003e\n  \u003csummary\u003e\u003cb\u003e🧠 Models \u0026 Architectures\u003c/b\u003e\u003c/summary\u003e\n  \n  - **Sequence Models:** RNN, LSTM, GRU, seq2seq with attention.\n  - **Transformers:** modern replacement for many sequence models; see the dedicated [Foundations: Transformers \u0026 LLMs](#foundations-transformers--llms) section for transformer specifics (self-attention, encoder/decoder variants, pretraining, PEFT, RAG, multilingual models).\n\u003c/details\u003e\n\n\u003cdetails\u003e\n  \u003csummary\u003e\u003cb\u003e🧰 Tooling \u0026 Libraries\u003c/b\u003e\u003c/summary\u003e\n  \n  - **Hugging Face Transformers \u0026 Datasets**: model hub, tokenizers, pipelines.\n  - **sentence-transformers**: semantic embeddings and retrieval.\n  - **spaCy, NLTK, Stanza, Flair**: preprocessing, tagging, parsing.\n  - **Vector DBs \u0026 Search:** FAISS, Chroma, Milvus, Pinecone.\n\u003c/details\u003e\n\n\u003cdetails\u003e\n  \u003csummary\u003e\u003cb\u003e📚 Resources\u003c/b\u003e\u003c/summary\u003e\n\n  - [Natural Language Processing Specialization, Deeplearning.ai](https://www.coursera.org/specializations/natural-language-processing)\n  - [DeepLearning.AI AI courses slides and resources](https://www.deeplearning.ai/resources/#course-slides)\n  - [RNN, LSTM and GRU in Arabic ](https://youtube.com/playlist?list=PLPBnj6azlABbMDuKRHn4rDJ69FRMXNaWk\u0026si=ctAUqELBzC4w3ApD)\n  - [Autoencoders in Arabic](https://youtube.com/playlist?list=PLPBnj6azlABZmpHXHm4MmcBaDNA3z80jc\u0026si=u9fBJxI6HL8GBMgI)\n  - Books: *Speech and Language Processing* (Jurafsky \u0026 Martin), *Natural Language Processing with Transformers*.\n  - Books: *Practical Natural Language Processing O'Reilly* (our college reference book)\n  - Books: *Natural Language Processing with Python O'Reilly* (our college reference book)\n\u003c/details\u003e\n\n### 2.4 Introduction to Computer Vision\n\nComputer Vision (CV) is the field of AI that enables machines to interpret and understand visual information from the world, such as images and videos. This section covers core concepts, tasks, models, tooling, and practical considerations for building CV systems.\n\u003cp style=\"color:gray;\"\u003eActually I am not a big fan of computer vision, so I will share with the basics that I have learned. If you want to learn more about it, you can check online courses and resources.\u003c/p\u003e\n\n\u003cdetails\u003e\n  \u003csummary\u003e\u003cb\u003e🔹 Core Concepts\u003c/b\u003e\u003c/summary\u003e\n\n  - **Image Representation:** Pixels, channels (RGB, grayscale), resolution, aspect ratio.\n  - **Feature Extraction:** Edges, corners, textures, shapes, keypoints.\n  - **Image Processing:** Filtering, convolution, morphological operations, color spaces (RGB, HSV, LAB).\n  - **Geometric Transformations:** Scaling, rotation, translation, affine transformations.\n  - **Image Enhancement:** Contrast adjustment, noise reduction, sharpening.\n\u003c/details\u003e\n\n\u003cdetails\u003e\n  \u003csummary\u003e\u003cb\u003e🔧 CV Tasks\u003c/b\u003e\u003c/summary\u003e\n\n  - **Image Classification:** Assigning labels to images (e.g., cat vs dog).\n  - **Object Detection:** Locating and classifying objects in images (bounding boxes).\n  - **Segmentation:** Pixel-level classification (semantic, instance, panoptic).\n  - **Image Generation:** Creating new images from scratch or editing existing ones.\n  - **Pose Estimation:** Detecting human or object poses and keypoints.\n  - **Optical Character Recognition (OCR):** Extracting text from images.\n  - **Face Recognition \u0026 Analysis:** Face detection, verification, emotion recognition.\n  - **Video Analysis:** Action recognition, tracking, scene understanding.\n\u003c/details\u003e\n\n\u003cdetails\u003e\n  \u003csummary\u003e\u003cb\u003e🧠 Models \u0026 Architectures\u003c/b\u003e\u003c/summary\u003e\n\n  - **Classical Methods:** Haar cascades, HOG + SVM, template matching.\n  - **Deep Learning Models:** CNNs, ResNet, EfficientNet, VGG, MobileNet, YOLO, SSD, Faster R-CNN.\n  - **Generative Models:** GANs, VAEs, Diffusion Models for image generation.\n  - **Vision Transformers (ViT):** Transformer-based architectures for vision tasks.\n  - **Foundation Models:** CLIP, DALL-E, Stable Diffusion.\n\u003c/details\u003e\n\n\u003cdetails\u003e\n  \u003csummary\u003e\u003cb\u003e🧰 Tooling \u0026 Libraries\u003c/b\u003e\u003c/summary\u003e\n\n  - **OpenCV:** Core library for image processing and computer vision tasks.\n  - **PyTorch \u0026 TensorFlow:** Deep learning frameworks with vision modules.\n  - **Hugging Face Transformers:** Pretrained models for vision tasks.\n  - **Scikit-Image:** Image processing in Python.\n\u003c/details\u003e\n\n\u003cdetails\u003e\n  \u003csummary\u003e\u003cb\u003e📚 Resources\u003c/b\u003e\u003c/summary\u003e\n\n  - [Computer Vision Dr.Ahmed Taha](https://youtube.com/playlist?list=PLPetKfNjOEGO0FECaXLNBDGRXDW4EGg7N\u0026si=GxGmMpu-3_hO7pYr)\n  - [DeepLearning.AI AI courses slides and resources](https://www.deeplearning.ai/resources/#course-slides)\n  - [AI topics including vision ](https://youtube.com/playlist?list=PL6-3IRz2XF5X-lzMZdmkvGAx1a3kIm7_I\u0026si=example)\n  - [My Kaggle Profile that has some image classification projects](https://www.kaggle.com/romanyn36)\n\u003c/details\u003e\n\n### 2.5 Data Preparation \nIn this section, you will learn how to convert raw, messy data into\nclean, structured, and model-ready data. Good data preparation often\nimproves your model more than changing the algorithm itself, this is most important topics to study.\n\u003cdetails open\u003e \n\u003csummary\u003e🧹 Data Cleaning (First Step Always)\u003c/summary\u003e\n\u003cul\u003e\n\u003cli\u003eHandling missing values (**drop**, **fill**, **interpolate**)\u003c/li\u003e\n\u003cli\u003eRemoving duplicates\u003c/li\u003e\n\u003cli\u003eFixing incorrect data types (string → number/date)\u003c/li\u003e\n\u003cli\u003eDetecting and treating outliers\u003c/li\u003e\n\u003cli\u003eCleaning inconsistent labels (e.g., Male/male/M)\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/details\u003e\n\n\u003cdetails\u003e \u003csummary\u003e🔄 Data Transformation\u003c/summary\u003e\n  \u003cul\u003e\n    \n  - Feature scaling: Normalization, **Standardization**\n  - Encoding categorical variables: **Label Encoding**, **One-Hot Encoding**\n  - Basic feature engineering: Creating new features, combining columns, feature selection\n  \u003c/ul\u003e\n\u003c/details\u003e\n\n\u003cdetails\u003e \u003csummary\u003e⚖️ Handling Imbalanced Data\u003c/summary\u003e\n\u003cul\u003e\n\u003cli\u003eOversampling\u003c/li\u003e\n\u003cli\u003eUndersampling\u003c/li\u003e\n\u003cli\u003eSMOTE (Synthetic sampling)\u003c/li\u003e\n\u003cli\u003eUsing class weights\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/details\u003e\n\n\u003cdetails\u003e \u003csummary\u003e📊 Data Splitting \u0026 Validation\u003c/summary\u003e\n\u003cul\u003e\n \n- Train / Validation / Test split\n- Stratified splitting\n- K-Fold Cross-Validation\n- Avoiding data leakage\n\u003c/ul\u003e\n\u003c/details\u003e\n\n\n\u003cdetails\u003e \u003csummary\u003e🎨 Data Augmentation\u003c/summary\u003e\n\u003cul\u003e\n\n- Image augmentation (**flip**, **rotate**, **crop**, **noise**)\n- Text augmentation (**synonyms**, **paraphrasing**)\n- Audio augmentation (**noise**, **pitch shift**, **time stretch**)\n\u003c/ul\u003e\n\u003c/details\u003e\n\n\u003cdetails\u003e \u003csummary\u003e💾 Working with Large Datasets\u003c/summary\u003e\n\u003cul\u003e\n\u003cli\u003eData generators / batch loading\u003c/li\u003e\n\u003cli\u003eEfficient formats (CSV)\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/details\u003e\n\n**Libraries**:**Pandas** \u0026 **NumPy**, **Scikit-learn**, **TensorFlow data loaders**\n  \n### 2.6 Data Science \u0026 Analysis \u0026 Visualization\n\nData Science involves extracting insights from data through analysis and visualization. This section covers the basics of exploring, analyzing, and visualizing data to inform decision-making.\nmost of the topics if reach here that means you are already familiar with them.\n\n\u003cdetails\u003e\n  \u003csummary\u003e\u003cb\u003e🔍 Exploratory Data Analysis (EDA)\u003c/b\u003e\u003c/summary\u003e\n  \n  - **Data Inspection:** Checking data types, missing values, duplicates, summary statistics.\n  - **Univariate Analysis:** Distributions, outliers, central tendency (mean, median, mode).\n  - **Bivariate/Multivariate Analysis:** Correlations, relationships between variables.\n  - **Data Profiling:** Understanding data structure and quality.\n  \n  \u003csummary\u003e\u003cb\u003eData Types \u0026 Categorical Variables\u003c/b\u003e\u003c/summary\u003e\n\n- **Numerical Data:** Continuous vs discrete, scaling, normalization.\n- **Categorical Data:** Nominal vs ordinal, encoding techniques (label, one-hot).\n- **Datetime Data:** Parsing, extracting features (year, month, day), handling time zones.\n\u003c/details\u003e \n\n\u003cdetails\u003e\n  \u003csummary\u003e\u003cb\u003e📈 Data Visualization\u003c/b\u003e\u003c/summary\u003e\n\n  - **Chart Types:** Bar charts, line plots, scatter plots, histograms, box plots, heatmaps.\n  - **Best Practices:** Choosing the right visualization, avoiding misleading charts, color theory.\n  - **Interactive Visualizations:** Basic dashboards with libraries like Plotly.\n\u003c/details\u003e\n\n\u003cdetails\u003e\n  \u003csummary\u003e\u003cb\u003e🧰 Tools \u0026 Libraries\u003c/b\u003e\u003c/summary\u003e\n\n  - **Pandas:** Data manipulation and analysis.\n  - **NumPy:** Numerical computing.\n  - **Matplotlib \u0026 Seaborn:** Static visualizations.\n  - **Plotly \u0026 Bokeh:** Interactive visualizations.\n  - **Jupyter Notebooks:** For exploratory analysis.\n\u003c/details\u003e\n\n\u003cdetails\u003e\n  \u003csummary\u003e\u003cb\u003e📚 Resources\u003c/b\u003e\u003c/summary\u003e\n  \n  - [Data Analysis with Python, Coursera](https://www.coursera.org/learn/data-analysis-with-python?specialization=ibm-data-science)\n  - [Kaggle profile](https://www.kaggle.com/romanyn36) (my projects and notebooks)\n  \n\u003c/details\u003e\n\n## Phase 3 – Agentic AI Systems\n### 3.1 Foundations: Transformers \u0026 LLMs\nUnderstand the core technology behind modern AI systems like ChatGPT, Gemini, Claude, and open-source LLMs.\n\n\u003cdetails\u003e\n  \u003csummary\u003e\u003cb\u003e🔹 Transformers — Basics\u003c/b\u003e\u003c/summary\u003e\n  \u003cul\u003e\n    \u003cli\u003eTokens \u0026 tokenization\u003c/li\u003e\n    \u003cli\u003eEmbeddings \u0026 positional encoding\u003c/li\u003e\n    \u003cli\u003eSelf-attention \u0026 multi-head attention\u003c/li\u003e\n    \u003cli\u003eEncoder vs Decoder architecture\u003c/li\u003e\n    \u003cli\u003eFeed-forward layers\u003c/li\u003e\n    \u003cli\u003eWhy Transformers replaced RNNs/CNNs in NLP\u003c/li\u003e\n    \u003cli\u003ePopular models: BERT, GPT, T5\u003c/li\u003e\n  \u003c/ul\u003e\n\u003c/details\u003e\n\n\u003cdetails\u003e\n  \u003csummary\u003e\u003cb\u003e🚀 Large Language Models (LLMs)\u003c/b\u003e\u003c/summary\u003e\n  \u003cul\u003e\n    \u003cli\u003ePretraining (next-token prediction)\u003c/li\u003e\n    \u003cli\u003eInstruction tuning\u003c/li\u003e\n    \u003cli\u003eRLHF / alignment techniques\u003c/li\u003e\n    \u003cli\u003eScaling laws \u0026 model sizes\u003c/li\u003e\n    \u003cli\u003eContext windows \u0026 long-context models\u003c/li\u003e\n    \u003cli\u003eInference basics (sampling, temperature, top-k/top-p)\u003c/li\u003e\n  \u003c/ul\u003e\n\u003c/details\u003e\n\n\u003cdetails\u003e\n  \u003csummary\u003e\u003cb\u003e⚙️ Fine-Tuning \u0026 Customization\u003c/b\u003e\u003c/summary\u003e\n  \u003cul\u003e\n    \u003cli\u003ePrompt engineering vs fine-tuning\u003c/li\u003e\n    \u003cli\u003eFull fine-tuning\u003c/li\u003e\n    \u003cli\u003eParameter-efficient methods (LoRA, QLoRA, adapters)\u003c/li\u003e\n    \u003cli\u003eInstruction / supervised fine-tuning (SFT)\u003c/li\u003e\n    \u003cli\u003eRetrieval-Augmented Generation (RAG)\u003c/li\u003e\n    \u003cli\u003eDomain adaptation \u0026 evaluation\u003c/li\u003e\n  \u003c/ul\u003e\n\u003c/details\u003e\n\n\u003cdetails\u003e\n  \u003csummary\u003e\u003cb\u003e🛠️ Applications \u0026 Practical Usage\u003c/b\u003e\u003c/summary\u003e\n  \u003cul\u003e\n    \u003cli\u003eChatbots \u0026 assistants\u003c/li\u003e\n    \u003cli\u003eRAG systems \u0026 knowledge search\u003c/li\u003e\n    \u003cli\u003eSummarization \u0026 question answering\u003c/li\u003e\n    \u003cli\u003eCode generation\u003c/li\u003e\n    \u003cli\u003eAgents \u0026 tool calling\u003c/li\u003e\n    \u003cli\u003eDeployment \u0026 optimization (quantization, batching, GPUs)\u003c/li\u003e\n    \u003cli\u003eOpen-source ecosystem: Hugging Face, vLLM, Ollama\u003c/li\u003e\n  \u003c/ul\u003e\n\u003c/details\u003e\n\n### 3.2 AI Agents: Concepts \u0026 Architectures\nLearn how to build intelligent systems that can reason, plan, use tools, and interact autonomously with users and environments.\n\n\u003cdetails\u003e\n  \u003csummary\u003e\u003cb\u003e📝 Prompt Engineering\u003c/b\u003e\u003c/summary\u003e\n  \u003cul\u003e\n    \u003cli\u003eUnderstanding prompts \u0026 instructions\u003c/li\u003e\n    \u003cli\u003ePrompt structure (system / user / tools)\u003c/li\u003e\n    \u003cli\u003eTypes of prompting: zero-shot, few-shot, chain-of-thought\u003c/li\u003e\n    \u003cli\u003eRole prompting \u0026 persona design\u003c/li\u003e\n    \u003cli\u003eOutput formatting (JSON, structured outputs)\u003c/li\u003e\n    \u003cli\u003ePrompt optimization \u0026 evaluation\u003c/li\u003e\n    \u003cli\u003eGuardrails \u0026 prompt safety\u003c/li\u003e\n  \u003c/ul\u003e\n\u003c/details\u003e\n\n\u003cdetails\u003e\n  \u003csummary\u003e\u003cb\u003e🧠 Agentic Concepts (Core Building Blocks)\u003c/b\u003e\u003c/summary\u003e\n  \u003cul\u003e\n    \u003cli\u003eEnvironment \u0026 context handling\u003c/li\u003e\n    \u003cli\u003eShort-term vs long-term memory\u003c/li\u003e\n    \u003cli\u003eMemory stores (vector DB, cache, databases)\u003c/li\u003e\n    \u003cli\u003ePersistence \u0026 state management\u003c/li\u003e\n    \u003cli\u003eTool / function calling/ enable-auto-tools choice in VLLMs and SGLang\u003c/li\u003e\n    \u003cli\u003ePlanning \u0026 reasoning loops\u003c/li\u003e\n    \u003cli\u003eReflection \u0026 self-correction\u003c/li\u003e\n    \u003cli\u003eSocial ability \u0026 human-in-the-loop\u003c/li\u003e\n  \u003c/ul\u003e\n\u003c/details\u003e\n\n\u003cdetails\u003e\n  \u003csummary\u003e\u003cb\u003e⚙️ Agent Architectures \u0026 Patterns\u003c/b\u003e\u003c/summary\u003e\n  \u003cul\u003e\n    \u003cli\u003eReactive agents\u003c/li\u003e\n    \u003cli\u003eDeliberative agents\u003c/li\u003e\n    \u003cli\u003eHybrid agents\u003c/li\u003e\n    \u003cli\u003eReAct (Reason + Act): \u003ca href=\"https://www.kaggle.com/code/romanyn36/react-agent-from-scratch\"\u003eReAct Agent from Scratch Kaggle\u003c/a\u003e, \u003ca href=\"https://github.com/romanyn36/RAG-Ai-Agent\"\u003eRAG (ReAct Based)\u003c/a\u003e, \u003ca href=\"https://github.com/romanyn36/california-procurement-agent\"\u003eAdvance ReAct Agent\u003c/a\u003e\u003c/li\u003e\n    \u003cli\u003ePlan-Execute\u003c/li\u003e\n    \u003cli\u003eTree/Graph of Thoughts\u003c/li\u003e\n    \u003cli\u003eTool-augmented agents\u003c/li\u003e\n    \u003cli\u003eWorkflow / pipeline agents\u003c/li\u003e\n    \u003cli\u003eMulti-agent collaboration systems (MAS, Crew-style)\u003c/li\u003e\n  \u003c/ul\u003e\n\u003c/details\u003e\n\n\u003cdetails\u003e\n  \u003csummary\u003e\u003cb\u003e📚 Knowledge \u0026 Retrieval (RAG)\u003c/b\u003e\u003c/summary\u003e\n  \u003cul\u003e\n    \u003cli\u003eDocument chunking strategies\u003c/li\u003e\n    \u003cli\u003eEmbeddings \u0026 vector databases\u003c/li\u003e\n    \u003cli\u003eSimilarity search\u003c/li\u003e\n    \u003cli\u003eRetrieval-Augmented Generation (RAG)\u003c/li\u003e\n    \u003cli\u003eHybrid search (keyword + vector)\u003c/li\u003e\n    \u003cli\u003eContext injection techniques\u003c/li\u003e\n  \u003c/ul\u003e\n\u003c/details\u003e\n\n\u003cdetails\u003e\n  \u003csummary\u003e\u003cb\u003e🛠️ Tools, Frameworks \u0026 Deployment\u003c/b\u003e\u003c/summary\u003e\n  \u003cul\u003e\n    \u003cli\u003eLangChain, LlamaIndex\u003c/li\u003e\n    \u003cli\u003eCrewAI / AutoGen / multi-agent frameworks\u003c/li\u003e\n    \u003cli\u003eOpenAI function calling / tool APIs\u003c/li\u003e\n    \u003cli\u003eFastAPI / backend integration\u003c/li\u003e\n    \u003cli\u003eAsync workflows \u0026 task queues\u003c/li\u003e\n    \u003cli\u003eMonitoring, logging, evaluation\u003c/li\u003e\n    \u003cli\u003eCost optimization \u0026 latency control\u003c/li\u003e\n  \u003c/ul\u003e\n\u003c/details\u003e\n\n\u003cdetails\u003e\n  \u003csummary\u003e🤖 AI agent frameworks\u003c/summary\u003e\n\u003cp style=\"color: green;\"\u003eThese are the most popular frameworks for building AI agents, and they are all open-source and free to use. I highly recommend learning at least one of them, as they will make your life much easier when building AI agents.\u003c/p\u003e\n  \u003cul\u003e\n\n- Langchain: for building AI agents with LLMs and tools :[LangChain Documentation](https://docs.langchain.com/oss/python/langchain/overview) \n- LangSmith: for monitoring and evaluating AI agents :[LangSmith Documentation](https://docs.langchain.com/langsmith/home)\n- Langgraph: for building AI agents with graph-based reasoning and agents orchestration :[Langgraph Documentation](https://docs.langchain.com/oss/python/langgraph/overview) - [AI Agents in Langgraph: DeepLearning.AI](https://learn.deeplearning.ai/courses/ai-agents-in-langgraph/), [Advanced AI Agents Tutorial: YouTube](https://youtu.be/xekw62yQu14?si=UWPOrljkjNaRiL0D) [AI Agents in Langgraph: GitHub](https://github.com/romanyn36/ai-agents-in-langgraph)\n- CrewAI: for multi-AI agent systems: [CrewAI Documentation](https://docs.crewai.com/), [Multi-AI Agent Systems with CrewAI: DeepLearning.AI](https://learn.deeplearning.ai/courses/multi-ai-agent-systems-with-crewai/), [CrewAI Crash Course: YouTube](https://youtu.be/sPzc6hMg7So?si=l8fd6lMoVDuKRhds), [How to use Ai Agents to do ALl your work: YouTube](https://youtu.be/ONKOXwucLvE?si=fuCZKyLdigZ5dLDT)\n  \n  \u003c/ul\u003e\n  \u003c/details\u003e\n- \u003cdetails\u003e\n  \u003csummary\u003e🖥️ LLMs Servers \u0026 APIs providers\u003c/summary\u003e\n  \u003cul\u003e\n\n- [OpenAI](https://developers.openai.com/api/docs)\n- [Google Generative AI](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/start/quickstart?usertype=adc)\n- [Google Ai studio \u0026 Vertex AI](https://docs.cloud.google.com/vertex-ai/docs)\n- [Groq](https://console.groq.com/docs/overview)\n- [Hugging Face](https://huggingface.co/)\n- [Anthropic](https://www.anthropic.com/)\n- [xAI](https://www.x.ai/)\n\n  \u003c/ul\u003e\n  \u003c/details\u003e\n\n  \u003cdetails\u003e\n  \u003csummary\u003e\u003cb\u003e🔒 Safety, Evaluation \u0026 Reliability\u003c/b\u003e\u003c/summary\u003e\n  \u003cul\u003e\n    \u003cli\u003eHallucination reduction (model temperature)\u003c/li\u003e\n    \u003cli\u003eOutput validation\u003c/li\u003e\n    \u003cli\u003eGuardrails \u0026 constraints\u003c/li\u003e\n    \u003cli\u003eAgent testing \u0026 benchmarking\u003c/li\u003e\n    \u003cli\u003eHuman feedback loops\u003c/li\u003e\n    \u003cli\u003eSecurity \u0026 prompt injection protection\u003c/li\u003e\n  \u003c/ul\u003e\n\u003c/details\u003e\n\n\n### 3.3 Building Agentic Systems (Tools \u0026 Infrastructure)\n- FastAPI: Building the API Layer for AI Agents: [Python FAST API Tutorial](https://youtu.be/-ykeT6kk4bk?si=XDAzehSEzrfbh4oC)\n- Chroma DB: The Ultimate Vector Database for AI Agents\n- MongoDB: The NoSQL Database for AI Agents, [MongoDB Essential Training LinkedIn Learning](https://www.linkedin.com/learning/mongodb-essential-training)\n- Redis: The In-Memory Data Store for AI Agents\n- RabbitMQ: Message Brokering for AI Agents\n\n### 3.4 Monitoring, Evaluation \u0026 Cost\n\n- Agent Monitoring \u0026 Evaluation\n- Evaluation Metrics: Measuring AI Agent Performance\n- Monitoring Tools: Keeping an Eye on AI Agents in Action\n- Cost Management: Optimizing Expenses for AI Agent Operations\n\n### 3.5 Deployment Production Ready Systems\n\n\u003cqoute\u003e (Saying Goodbye to Localhost and welcoming the World) \u003c/qoute\u003e\n\n\u003cdetails\u003e\n  \u003csummary\u003e\u003cb\u003e⚙️ Deployment Optimization\u003c/b\u003e\u003c/summary\u003e\n  \u003cul\u003e\n    \u003cli\u003e\u003cb\u003eInference:\u003c/b\u003e batching, streaming, caching, quantization (int8, 4-bit),fp16\u003c/li\u003e\n    \u003cli\u003e\u003cb\u003eLatency \u0026 Cost:\u003c/b\u003e model size vs latency trade-offs, distillation, pruning.\u003c/li\u003e\n    \u003cli\u003e\u003cb\u003eSafety:\u003c/b\u003e input validation, hallucination mitigation, output filters, user feedback loops.\u003c/li\u003e\n  \u003c/ul\u003e\n\u003c/details\u003e\n\n\u003cdetails\u003e\n  \u003csummary\u003e\u003cb\u003e🐳 Containerization \u0026 Orchestration\u003c/b\u003e\u003c/summary\u003e\n\n  - **Docker**: Containerizing AI Agents for Scalable Deployment, [Docker Foundations Professional Certificate](https://www.linkedin.com/learning/paths/docker-foundations-professional-certificate)\u003c/li\u003e\n  - **Nginx**:Load Balancing and Reverse Proxy for AI Agent APIs (if building your own API from scratch without using a service).\u003c/li\u003e\n  \u003c/ul\u003e\n\u003c/details\u003e\n\n\u003cdetails\u003e\n  \u003csummary\u003e\u003cb\u003e🚀 High-Performance Inference Engines\u003c/b\u003e\u003c/summary\u003e\n  \u003cul\u003e\n\n  - **VLLM \u0026 SGLang**: High-Performance Inference Engines for AI Agents - [VLLM Docs](https://docs.vllm.ai/en/latest/), [SGLang Docs](https://docs.sglang.io/)\n    - **Custom Inference Templates:** Enabling auto-tool choice in VLLM ([Tool Calling](https://docs.vllm.ai/en/latest/features/tool_calling/)) and SGLang.\n  \u003c/ul\u003e\n\u003c/details\u003e\n\n\u003cdetails\u003e\n  \u003csummary\u003e\u003cb\u003e☁️ Cloud Platforms \u0026 GPU Providers\u003c/b\u003e\u003c/summary\u003e\n  \u003cp\u003e\u003cem\u003eactually there are thousands of cloud providers, but these are what I've used over the past 3 years and recommend.\u003c/em\u003e\u003c/p\u003e\n  \u003cul\u003e\n    \u003cli\u003e\u003cb\u003eGCP:\u003c/b\u003e Google Cloud Platform and its vertex AI services(Google Generative AI,model garden for open-source models)\u003c/li\u003e\n    \u003cli\u003e\u003cb\u003eDigital Ocean:\u003c/b\u003ei use its vps for hosting small projects and APIs\u003c/li\u003e\n    \u003cli\u003e\u003cb\u003eHeroku:\u003c/b\u003e Platform as a Service, it's good for hosting small projects\u003c/li\u003e\n    \u003cli\u003e\u003cb\u003eHosting.com \u0026 Hostinger:\u003c/b\u003e the cheapest hosting providers.\u003c/li\u003e\n    \u003cli\u003e\u003cb\u003eVast.ai \u0026 RunPod:\u003c/b\u003e GPU cloud providers when you need serve your LLM or any other models\u003c/li\u003e\n    \u003cp\u003e\u003cem\u003eBut please, when dealing with hosting, keep your eyes on 3 things: \u003cspan style=\"color:red;\"\u003e1. your visa 😭😭\u003c/span\u003e, 2. security and firewalls, 3. resource usage.\u003c/span\u003e\u003c/em\u003e\u003c/p\u003e\n  \u003c/ul\u003e\n\u003c/details\u003e\n\n- GPUs: Accelerating AI Agents with Graphics Processing Units\n\n---\n\n## 📂 Additional Resources\n\n🌟 **Open to contributions** — feel free to suggest improvements or additional resources!\n\nThis repository includes additional files with extra learning materials:\n- 📱 [`ACCOUNTS_TO_FOLLOW.md`](ACCOUNTS_TO_FOLLOW.md) - Recommended social media accounts for AI/ML content\n- 📺 [`CHANNELS_TO_FOLLOW.md`](CHANNELS_TO_FOLLOW.md) - YouTube channels and content creators\n- 📚 [`EXTRA_RESOURCES.md`](EXTRA_RESOURCES.md) - Additional learning materials and resources\n- 💡 [`PROJECTS_IDEAS.md`](PROJECTS_IDEAS.md) - Practical project ideas to apply your learning\n\n\u003e ⚠️ **Note:** Resources in these additional files may include community contributions and materials I haven't personally verified or watched yet. The main roadmap above contains only resources I've personally used and recommend.\n\n---\n\n## 🤝 Contributing\n\nContributions are welcome! If you have suggestions, resources, or improvements:\n\n1. 🍴 Fork the repository\n2. 🔨 Create your feature branch (`git checkout -b feature/AmazingResource`)\n3. 💾 Commit your changes (`git commit -m 'Add some AmazingResource'`)\n4. 📤 Push to the branch (`git push origin feature/AmazingResource`)\n5. 🎯 Open a Pull Request\n\n---\n\n## ⭐ Show Your Support\n\nIf this roadmap helped you in your AI learning journey, please consider:\n\n- ⭐ **Starring this repository**\n- 🔄 **Sharing it with others**\n- 💬 **Providing feedback or suggestions**\n\n---\n\n## 👨‍💻 About Me\n\nBefore we dive in — I'm Romani. Been freelancing since 2023, now building AI systems at **Penny Software**. Navigating the overwhelming world of AI learning pushed me to create this roadmap: a clear, practical path built on hands-on projects and verified resources, not endless theory.\n\n\u003ctable\u003e\n\u003ctr\u003e\n\u003ctd width=\"200\"\u003e\n\u003cimg src=\"https://upload.wikimedia.org/wikipedia/en/e/e7/Steve_%28Minecraft%29.png?20220417165835\" width=\"150\" alt=\"Romani Nasrat\"/\u003e\n\u003c/td\u003e\n\u003ctd\u003e\n\n### Romani Nasrat\n**AI Engineer @ Penny Software | AI/ML \u0026 Agentic AI Specialist**\n\n🎓 Graduate of **Faculty of Computers and Artificial Intelligence**, Benha University  \n💼 AI Engineer at **Penny Software** - Building production-grade AI solutions  \n🤖 Specializing in **LLMs, AI Agents, RAG Pipelines, Vector Databases**  \n⚡ Building scalable agentic systems with **LangChain, CrewAI, PyTorch, FastAPI, Django**  \n🌐 Portfolio: **[romaninasrat.com](http://romaninasrat.com/)**  \n🎻 Violin player | ♟️ Chess enthusiast | 🎮 Minecraft Expert haha\n\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003c/table\u003e\n\n---\n\n## 📬 Contact \u0026 Connect\n\n\u003cdiv align=\"center\"\u003e\n\n**Have questions? Want to collaborate? Let's connect!**\n\n[![Portfolio](https://img.shields.io/badge/Portfolio-FF5722?style=for-the-badge\u0026logo=google-chrome\u0026logoColor=white)](http://romaninasrat.com/)\n[![LinkedIn](https://img.shields.io/badge/LinkedIn-0A66C2?style=for-the-badge\u0026logo=linkedin\u0026logoColor=white)](https://www.linkedin.com/in/romaninasrat/)\n[![GitHub](https://img.shields.io/badge/GitHub-181717?style=for-the-badge\u0026logo=github\u0026logoColor=white)](https://github.com/romanyn36)\n[![Twitter](https://img.shields.io/badge/Twitter-1DA1F2?style=for-the-badge\u0026logo=twitter\u0026logoColor=white)](https://x.com/RomaniNasrat)\n[![Kaggle](https://img.shields.io/badge/Kaggle-20BEFF?style=for-the-badge\u0026logo=kaggle\u0026logoColor=white)](https://kaggle.com/romanyn36)\n[![Telegram](https://img.shields.io/badge/Telegram-26A5E4?style=for-the-badge\u0026logo=telegram\u0026logoColor=white)](https://t.me/romanyn36)\n[![Email](https://img.shields.io/badge/Email_Me-EA4335?style=for-the-badge\u0026logo=gmail\u0026logoColor=white)](mailto:romani.nasrat@gmail.com)\n\n\u003cbr/\u003e\n\n**Made with ❤️ by [Romani Nasrat](https://github.com/romanyn36)**\n\n© 2026 Agentic AI Roadmap • [MIT License](LICENSE)\n\n\u003c/div\u003e\n\n---\n\n\u003cdiv align=\"center\"\u003e\n  \u003ca href=\"#-the-complete-ai-engineer-roadmap\"\u003e⬆️ Back to Top\u003c/a\u003e\n\u003c/div\u003e\n\n\n\n\n\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fromanyn36%2Fagentic-ai-roadmap","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fromanyn36%2Fagentic-ai-roadmap","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fromanyn36%2Fagentic-ai-roadmap/lists"}