Nadeem S.

I'm a Senior AI Integration Engineer and Python/Django/FastAPI developer with 10+ years of experience building production backend systems and AI-powered applications. Over the last two years, I've specialized in integrating LLMs (Claude, OpenAI GPT, Gemini, open-source models) into real business products systems that handle real users, real data, and real scale. If you want someone to integrate an LLM, build a RAG pipeline, or set up context engineering for your product and actually ship it to production, that's my core focus. I handle the full stack: the LLM layer, retrieval and context design, vector store, Django/FastAPI backend, auth, deployment, and monitoring. What I specialize in: *LLM & AI Integration* Claude (Anthropic) — Messages API, tool use, computer use, MCP (Model Context Protocol), extended thinking OpenAI — GPT-4/5, Assistants API, Responses API, function calling, structured outputs Google Gemini, Mistral, Groq, open-source models (Llama, Qwen) via Ollama / vLLM Multi-provider routing, fallback logic, and cost optimization *RAG & Context Engineering* Production RAG pipelines: ingestion, chunking strategies, hybrid search (BM25 + vector), reranking Context engineering, designing how the right information gets into the model's context window at the right time Agentic RAG, query rewriting, self-querying retrievers, and evaluation frameworks (Ragas, DeepEval) Document parsing for complex PDFs, tables, images (Unstructured, LlamaParse, Docling) *AI Agents & Chatbots* Agentic workflows with tool use, planning, memory, and multi-step reasoning Custom chatbots with persistent memory, streaming responses, and human-in-the-loop flows LangChain, LangGraph, LlamaIndex, CrewAI, Pydantic AI MCP server development to expose custom tools to Claude and other agents *Vector Databases* Pinecone, Qdrant, Weaviate, Chroma, pgvector, Milvus — including index design, metadata filtering, and hybrid search. *Python Backend (where the AI actually lives)* Django, Django REST Framework, FastAPI, Flask Async Python, streaming endpoints (SSE, WebSockets) for real-time LLM responses Celery, Redis, RabbitMQ for background AI jobs GraphQL (Strawberry, Graphene) *Databases* PostgreSQL, MySQL, MongoDB, Redis, Firestore *Cloud & DevOps* AWS (Lambda, EC2, ECS, Bedrock, S3, RDS, API Gateway), GCP (Vertex AI), Azure OpenAI Docker, Docker Compose, CI/CD pipelines Nginx, Gunicorn, uWSGI, monitoring & observability (Langfuse, LangSmith, Helicone) *Automation & Scraping* Playwright, Selenium, BeautifulSoup, custom automation bots (checkout, ticketing, inventory, lead scraping, LinkedIn automation). *Integrations* Stripe, third-party REST/webhook integrations, Firebase, Twilio, SendGrid, Swagger/Redoc API documentation. *Testing & Quality* Pytest, TDD, LLM evaluation pipelines, prompt regression testing. What I deliver: AI features that actually work in production, not just in the happy path Clean Django/FastAPI backends with proper architecture RAG systems tuned and evaluated for accuracy, not just "plugged in" Cost-aware LLM usage (caching, batching, smaller models where possible) Secure deployments, monitoring, and long-term maintenance If you're building an AI product, modernizing an existing Python/Django codebase with LLM features, or need someone who can own both the AI layer and the backend it sits on, let's talk.