{"id":15442424,"url":"https://github.com/ricoledan/deloitte-insightbot","last_synced_at":"2026-05-17T04:38:59.672Z","repository":{"id":256299058,"uuid":"854626218","full_name":"Ricoledan/deloitte-insightbot","owner":"Ricoledan","description":"💬 Retrieval Augmented Generation (RAG) system based on Deloitte's Weekly Global Economic Update ","archived":false,"fork":false,"pushed_at":"2024-09-13T16:16:12.000Z","size":20,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-03-28T07:13:18.536Z","etag":null,"topics":["chromadb","deloitte","langchain","llm","python","retrieval-augmented-generation"],"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/Ricoledan.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}},"created_at":"2024-09-09T14:00:52.000Z","updated_at":"2024-09-13T16:16:16.000Z","dependencies_parsed_at":"2024-09-13T10:56:51.413Z","dependency_job_id":"2fc7edb1-59d0-489d-bbb7-a172b09b906c","html_url":"https://github.com/Ricoledan/deloitte-insightbot","commit_stats":null,"previous_names":["ricoledan/deloitte-insightbot"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Ricoledan%2Fdeloitte-insightbot","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Ricoledan%2Fdeloitte-insightbot/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Ricoledan%2Fdeloitte-insightbot/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Ricoledan%2Fdeloitte-insightbot/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/Ricoledan","download_url":"https://codeload.github.com/Ricoledan/deloitte-insightbot/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":245984590,"owners_count":20704798,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","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":["chromadb","deloitte","langchain","llm","python","retrieval-augmented-generation"],"created_at":"2024-10-01T19:27:26.149Z","updated_at":"2026-05-17T04:38:59.643Z","avatar_url":"https://github.com/Ricoledan.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Deloitte-Insightbot\n\n## Overview\n\nThe `deloitte-insightbot` is a question and answer system designed to provide insights based on Deloitte's weekly\neconomic updates. These updates offer a brief overview of the global political and economic situation, summarizing key\nimpacts and trends.\n\n## Features\n\n- **Data Ingestion**: Fetches content from Deloitte's weekly economic\n  update [URL](https://www2.deloitte.com/us/en/insights/economy/global-economic-outlook/weekly-update/weekly-update-2023-10.html?icid=archive_click).\n- **Embeddings Storage**: Stores embeddings of the content in a VectorDB.\n- **Retrieval-Augmented Generation**: Retrieves relevant passages to generate answers for user queries using an LLM.\n\n## Components\n\n- **Data Ingestion**: A module to scrape and parse content from the specified URL.\n    - `UnstructuredURLLoader` class to fetch and parse the content from the URL.\n- **Embeddings Model**: Utilizes an embedding model to convert content into vector representations.\n    - `OpenAIEmbeddings` model with the model name `text-embedding-3-large`.\n- **VectorDB**: Stores the embeddings for efficient retrieval.\n    - `Chroma` class from langchain_chroma is used to interact with ChromaDB.\n- **LLM**: Generates answers based on the retrieved passages.\n    - ChatOpenAI class with the model name `gpt-3.5-turbo`.\n\n## Usage\n\n1. **Ingest Data**: Run the data ingestion script to fetch and parse the content.\n2. **Store Embeddings**: Use the embeddings model to convert the content into vectors and store them in the VectorDB.\n3. **Query System**: Input a user query to retrieve relevant passages and generate an answer using the LLM.\n\n## Commands\n\nInstall the required packages\n\n```bash\npip install -r requirements.txt\n```\n\nStart the ChromaDB container\n\n```bash\ndocker compose up -d\n```\n\nPing the ChromaDB container to check if it is running\n\n```bash\ncurl localhost:8000/api/v1/heartbeat\n```\n\nRun the application\n\n```bash\npython src/main.py\n```","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fricoledan%2Fdeloitte-insightbot","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fricoledan%2Fdeloitte-insightbot","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fricoledan%2Fdeloitte-insightbot/lists"}