{"id":39037569,"url":"https://github.com/lukebrady/ai-engineering","last_synced_at":"2026-01-17T17:45:22.433Z","repository":{"id":310375882,"uuid":"1039593463","full_name":"lukebrady/ai-engineering","owner":"lukebrady","description":"AI Engineering in the open","archived":false,"fork":false,"pushed_at":"2025-09-26T04:28:14.000Z","size":375,"stargazers_count":0,"open_issues_count":2,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2025-09-26T05:46:17.997Z","etag":null,"topics":["ai","ai-agents","ai-engineering"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/lukebrady.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"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":"AGENTS.md","dco":null,"cla":null}},"created_at":"2025-08-17T15:18:16.000Z","updated_at":"2025-09-26T03:59:25.000Z","dependencies_parsed_at":null,"dependency_job_id":"60e1ae48-27fd-4684-ad94-7cdd1f95b060","html_url":"https://github.com/lukebrady/ai-engineering","commit_stats":null,"previous_names":["lukebrady/ai-engineering"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/lukebrady/ai-engineering","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/lukebrady%2Fai-engineering","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/lukebrady%2Fai-engineering/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/lukebrady%2Fai-engineering/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/lukebrady%2Fai-engineering/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/lukebrady","download_url":"https://codeload.github.com/lukebrady/ai-engineering/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/lukebrady%2Fai-engineering/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":28513975,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-01-17T13:38:16.342Z","status":"ssl_error","status_checked_at":"2026-01-17T13:37:44.060Z","response_time":85,"last_error":"SSL_read: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"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":["ai","ai-agents","ai-engineering"],"created_at":"2026-01-17T17:45:22.298Z","updated_at":"2026-01-17T17:45:22.404Z","avatar_url":"https://github.com/lukebrady.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# AI Engineering\n\nA comprehensive collection of AI engineering projects designed as an **executable codebase** - where both humans and AI agents can discover and run available tools through a unified interface. This repository demonstrates practical implementations across cloud infrastructure, AI inference systems, and graph-based AI architectures.\n\n[![OpenTofu](https://img.shields.io/badge/OpenTofu-1.0+-blue.svg)](https://opentofu.org/)\n[![Packer](https://img.shields.io/badge/Packer-1.7+-green.svg)](https://packer.io/)\n[![AWS](https://img.shields.io/badge/AWS-EC2%20%7C%20EBS-orange.svg)](https://aws.amazon.com/)\n\n## 🚀 Quick Start\n\n```bash\n# Get available commands and project overview\nmake help\n\n# Check prerequisites (Packer, AWS CLI, OpenTofu)\nmake check-prerequisites\n\n# Initialize and setup the project\nmake setup\n```\n\n## 📁 Project Structure\n\n```text\nai-engineering/\n├── Makefile                    # 🎯 EXECUTABLE INTERFACE - Universal tool catalog\n├── README.md                   # This file\n├── .github/workflows/          # CI/CD automation for tool validation\n├── agents/                     # AI agent implementations and examples\n│   └── oss-agent/             # Example agent using open source models\n├── infrastructure/             # Cloud infrastructure projects\n│   └── ai-inference/           # Production-ready AI inference infrastructure\n│       ├── packer/             # Custom Ubuntu 24.04 AMI builder with GPU support\n│       └── opentofu/           # Modular Infrastructure as Code deployment\n│           ├── iam/            # IAM roles and permissions\n│           ├── inference/      # Main deployment using modular components\n│           └── modules/        # Reusable Terraform modules\n│               └── inference/  # Parameterized inference server module\n└── intro-langgraph/            # (Planned) LangGraph learning project\n```\n\n## 🎯 Project Philosophy: Executable Codebase\n\nThis repository embodies a core principle: **the codebase itself should be executable by both humans and AI agents**. Every tool, command, and capability is discoverable and runnable through a unified interface.\n\n### For AI Agents 🤖\n\n```bash\nmake help  # Same interface - AI agents can parse and execute tools\n```\n\n**Key Principles:**\n\n- **Single Source of Truth**: The Makefile serves as the authoritative catalog of all executable operations\n- **Self-Documenting**: Every command includes clear descriptions and usage examples\n- **Universal Access**: The same interface works for humans, CI/CD systems, and AI agents\n- **Discoverability**: No hidden commands - everything is accessible via `make help`\n\n## 🎯 Project Goals\n\nThis repository serves as a comprehensive portfolio demonstrating:\n\n- **Infrastructure as Code**: Automated cloud infrastructure provisioning\n- **AI Inference Systems**: Production-ready AI model deployment  \n- **DevOps Best Practices**: CI/CD, automation, and monitoring\n- **Human-AI Collaboration**: Interfaces designed for both human and AI agent interaction\n- **Learning in Public**: Documented journey through AI engineering\n\n## 🏗️ Current Projects\n\n### 1. AI Inference Infrastructure (`infrastructure/ai-inference/`)\n\nA complete infrastructure solution for deploying AI inference workloads on AWS using custom AMIs and modular OpenTofu configuration.\n\n**Key Features:**\n\n- Custom Ubuntu 24.04 AMI with Docker, NVIDIA drivers, and GPU support\n- Modular architecture: separate IAM, inference, and reusable modules\n- Multi-model deployment: Qwen 3 0.6B, GPT-OSS 20B, and Gemma 3 27B configurations\n- vLLM server with systemd integration and container lifecycle management\n- GPU-enabled instances (g5.2xlarge) with automated provisioning\n- Comprehensive security: EBS encryption, restrictive security groups, IAM best practices\n\n**Quick Deploy:**\n\n```bash\n# Build custom AMI\nmake ami-build\n\n# Deploy IAM resources\nmake tofu-iam-apply\n\n# Deploy inference infrastructure\nmake tofu-inference-apply\n```\n\n### 2. OSS Agent Example (`agents/oss-agent/`)\n\nA practical example of building AI agents using open source models with comprehensive documentation.\n\n**Key Features:**\n\n- OpenAI-compatible API integration for local models (vLLM, Ollama)\n- Function calling with Wikipedia search tools\n- Interactive REPL with Rich formatting\n- Comprehensive comments explaining agent patterns and OSS model usage\n\n**Quick Start:**\n\n```bash\ncd agents/oss-agent\npython main.py  # Interactive agent with Wikipedia search\n```\n\n## 🛠️ Technologies\n\n- **Infrastructure**: OpenTofu (Terraform), Packer, AWS (EC2, EBS, VPC, IAM)\n- **AI/ML**: vLLM, Python, GPU acceleration, OpenAI-compatible APIs\n- **Containerization**: Docker, systemd service management\n- **Automation**: Make, Bash scripting, GitHub Actions\n- **Security**: EBS encryption, security groups, IAM best practices\n\n## 📋 Executable Tool Catalog\n\nThe project's **Makefile serves as the executable interface** - a programmatically parseable catalog of all available tools. This design enables both humans and AI agents to discover and execute operations using the same commands.\n\n### 🔍 Tool Discovery\n\n```bash\nmake help  # Lists all available tools with descriptions\n```\n\n**Example Output:**\n\n```text\nAI Engineering - Executable Tool Catalog\n\nAMI/Packer Commands:\n  ami-build                 Build the AMI (with validation)\n  ami-init                  Initialize Packer plugins\n  ami-validate              Validate Packer configuration\n\nAgent Commands:\n  agent-oss-check           Check OSS agent environment and dependencies\n  agent-oss-install         Install OSS agent dependencies\n  agent-oss-run             Run the OSS agent interactively\n\nInfrastructure/OpenTofu Commands:\n  tofu-apply                Deploy all infrastructure with OpenTofu\n  tofu-iam-apply            Deploy IAM infrastructure with OpenTofu\n  tofu-inference-apply      Deploy inference infrastructure with OpenTofu\n  tofu-init                 Initialize all OpenTofu modules\n  tofu-plan                 Show deployment plan for all modules\n  tofu-validate             Validate all OpenTofu modules\n\nSetup \u0026 Verification Commands:\n  check-aws-config          Check AWS configuration and permissions\n  check-prerequisites       Check if required tools are installed\n  setup                     Complete setup and initialization\n\nUtility Commands:\n  help                      Show this help message\n  list-ami                  List recent AMIs created by this project\n  status                    Show current project status\n```\n\n### 🤖 AI Agent Compatibility\n\nThe Makefile format is specifically designed to be:\n\n- **Parseable**: AI agents can extract command names and descriptions\n- **Executable**: Commands can be run programmatically\n- **Self-Contained**: Each command includes all necessary context\n- **Consistent**: Uniform pattern across all operations\n\n### 📊 Tool Categories\n\n**Infrastructure Commands:**\n\n- `make ami-build` - Build custom Ubuntu AI inference AMI\n- `make tofu-iam-apply` - Deploy IAM resources (roles, policies)\n- `make tofu-inference-apply` - Deploy inference infrastructure (EC2, vLLM)\n- `make tofu-apply` - Deploy all infrastructure (IAM + inference)\n- `make tofu-destroy` - Tear down all infrastructure\n\n**Agent Commands:**\n\n- `make agent-oss-run` - Run the OSS agent interactively\n- `make agent-oss-install` - Install OSS agent dependencies\n- `make agent-oss-check` - Check agent environment and dependencies\n\n**Utility Commands:**\n\n- `make check-prerequisites` - Verify required tools\n- `make setup` - Complete project initialization\n- `make status` - Show current project status\n- `make clean` - Clean build artifacts\n\n**Validation Commands:**\n\n- `make ami-validate` - Validate Packer configuration\n- `make tofu-validate` - Validate OpenTofu configuration\n\n## 🚧 Planned Projects\n\n### 2. LangGraph Introduction (`intro-langgraph/`)\n\n- Graph-based AI application development\n- Multi-agent systems and workflows\n- Integration with various LLM providers\n\n### 3. Additional Infrastructure Components\n\n- Container orchestration (EKS)\n- Model serving platforms\n- Monitoring and observability stack\n\n## 🔧 Prerequisites\n\nBefore using this project, ensure you have:\n\n1. **AWS CLI** configured with appropriate permissions\n2. **Packer** 1.7+ for AMI building\n3. **OpenTofu** 1.0+ for infrastructure deployment\n4. **Make** for automation commands\n\nQuick installation on macOS:\n\n```bash\nbrew install awscli packer opentofu\n```\n\n## 📖 Documentation\n\nEach subproject contains detailed documentation:\n\n- [`infrastructure/ai-inference/packer/README.md`](infrastructure/ai-inference/packer/README.md) - AMI building guide\n- [`infrastructure/ai-inference/opentofu/README.md`](infrastructure/ai-inference/opentofu/README.md) - Complete infrastructure deployment guide\n- [`agents/oss-agent/main.py`](agents/oss-agent/main.py) - Comprehensive agent implementation with inline documentation\n\n## 🤝 Contributing\n\nThis is a personal learning repository, but feedback and suggestions are welcome! Please:\n\n1. Check existing issues and documentation\n2. Open an issue for bugs or feature requests  \n3. Follow the established code and documentation patterns\n\n## 🎓 Learning Resources\n\nThis repository represents practical implementations learned from:\n\n- AWS Well-Architected Framework\n- Infrastructure as Code best practices\n- AI/ML deployment patterns\n- DevOps automation principles\n\n---\n\n*This project demonstrates production-ready infrastructure automation, AI system deployment, and continuous learning in the rapidly evolving field of AI engineering.*\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Flukebrady%2Fai-engineering","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Flukebrady%2Fai-engineering","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Flukebrady%2Fai-engineering/lists"}