Skip to content
View bdeva1975's full-sized avatar

Block or report bdeva1975

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
bdeva1975/README.md

Hi, I'm Devasish Banerjee 👋

Senior Manager — GenAI & Application Development TCG Lifesciences Pvt. Ltd. | Kolkata, India 🇮🇳


🧠 What I Do

I build AI-powered applications and open source tools at the intersection of Generative AI, pharma/CRO workflows, and enterprise application development.

  • 🔬 Leading GenAI initiatives for 1,300+ scientists at a pharma CRO
  • 🤖 Building multi-agent AI systems for drug discovery workflows
  • 🛠️ Full-stack developer — Spring Boot · React · PostgreSQL · Python
  • 📦 Publishing open source AI tools on GitHub and PyPI

🚀 Featured Open Source Projects

Project Description Stack
LLM Cost Autopilot LLM FinOps control loop — observe → analyze → detect → predict → recommend → simulate → govern → automate. YAML policy rules, conditioned approvals with revalidation, anomaly detection & cost simulation. 135 tests, CI green Python · Streamlit
Model Regression Detector Catch model performance regressions before they ship — classification & regression tasks, deterministic offline evaluation, methodology docs Python · Streamlit · scikit-learn
AgentGov Governance & observability control plane for agentic systems — policy/budget enforcement, audit trail, MCP integration Python · FastAPI · Streamlit
ChemMCP FastMCP-based chemistry MCP server exposing cheminformatics tools to LLM agents Python · FastMCP · RDKit
HallucinationBench Detect hallucinations in RAG pipeline output in 2 lines of Python Python · OpenAI · Streamlit
rag-indexing-benchmark Compare 6 RAG indexing strategies on your own documents — with a single command Python · LangChain

🧰 Tech Stack

AI/ML: OpenAI API · Claude API · LangChain · LangGraph · Streamlit · RAG · Multi-agent systems · MCP Backend: Spring Boot 3.x · Java · Python · REST APIs Frontend: React 19 · TypeScript · Tailwind CSS Database: PostgreSQL · Oracle DevOps: AWS ECS · Docker · GitHub Actions · uv Domain: Pharma CRO · Drug Discovery · DMPK · Regulatory workflows


📊 GitHub Stats

Devasish's GitHub Stats Top Languages


📦 Latest PyPI Release

HallucinationBench

pip install hallucinationbench

🌐 Connect With Me

LinkedIn GitHub PyPI


Currently building open source AI governance, evaluation, and agent infrastructure tools. Open to consulting, collaboration, and speaking opportunities.

Popular repositories Loading

  1. SummaryForge SummaryForge Public

    SummaryForge: An AI-powered PDF summarization tool built with Python and Streamlit. Upload any PDF, specify your summarization needs, and get instant, customized summaries using OpenAI's GPT models…

    Python 2

  2. hallucinationbench hallucinationbench Public

    Detect hallucinations in your RAG pipeline output — in two lines of Python.

    Python 2

  3. prompt2pic prompt2pic Public

    AI-powered image generator using OpenAI's DALL-E. Transform text prompts into stunning visuals with ease. Features include negative prompt support, customizable image sizes, and simple integration.…

    Python 1

  4. FlowMinds FlowMinds Public

    FlowMinds is an AI-powered web application that generates detailed flowcharts from natural language input. Built with Streamlit and OpenAI's GPT model, it transforms complex processes into visual r…

    Python 1

  5. llm-research-engine llm-research-engine Public

    LangChain-based web research engine with async scraping and multi-backend search

    Python 1

  6. rag-indexing-benchmark rag-indexing-benchmark Public

    Compare 6 RAG indexing strategies on your own documents — with a single command

    Python 1 2