Crafting intelligent, scalable, and low-latency AI applications.
Passionate about LLMs, Hybrid Retrieval-Augmented Generation (RAG), and Autonomous Multi-Agent Workflows.
- 🔭 Currently Building: production-rag-engine — An enterprise-grade Hybrid RAG system with Dense + BM25 search, Cross-Encoder reranking, and automated RAGAS evaluation.
- 💡 Core Interests: Production LLMOps, Vector Databases, Fast Inference Architectures, Agentic Workflows (LangGraph).
- 🌱 Learning & Researching: Advanced Context Optimization, Small Language Models (SLMs) fine-tuning, and LLM-as-a-Judge Evaluation.
- 💬 Ask me about: Python, RAG Architecture, FastAPI, Groq Inference, and Vector Embeddings.
| Domain | Technologies & Frameworks |
|---|---|
| 🤖 LLMs & Agents | LangChain, LangGraph, Groq LPU, OpenAI, Google Gemini, Ollama |
| 🔍 Search & Vector DBs | ChromaDB, Qdrant, BM25 Keyword Search, FlashRank Reranker |
| 📊 Evaluation & Ops | RAGAS, Pytest, Docker, GitHub Actions (CI/CD) |
| ⚡ Backend & Serving | Python 3.11+, FastAPI, Pydantic v2, Uvicorn, Streamlit |
|
Enterprise-ready Hybrid RAG Pipeline powered by Groq LPU (Llama 3.3), ChromaDB, BM25, FlashRank Cross-Encoder, and Automated RAGAS Evaluation.
|
Designed & built with 💻 by Daffa Arigoh. Let's connect and build the future of AI together!