I build production-grade AI agents and LLM-powered automation systems that actually ship. Drawing on real-world experience deploying autonomous agents into live infrastructure, I help founders and small teams go from "we should automate this" to working code in days, not months.
What I deliver:
- Custom AI agents with real tool use, function calling, and multi-step reasoning
- LLM orchestration across providers (Groq, Claude, OpenAI) with fallback and routing logic
- Data pipelines: web scrapers, ETL workflows, transaction reconciliation, scoring engines
- ML systems: LightGBM/XGBoost ensembles, feature engineering, model training pipelines
- Cloud infrastructure on Oracle Cloud (OCI), AWS, or your platform of choice
- MCP (Model Context Protocol) server integration for Claude Desktop and similar clients
Tech stack: Python, FastAPI, Groq API, Anthropic Claude API, OpenAI API, MCP SDKs, LangChain, Pandas, OpenPyXL, BeautifulSoup, asyncio, SSH/Linux server admin, OCI compute and storage, SQLite/Postgres, REST and webhook integrations.
Why work with me:
- I run my own production AI platform - I've shipped exactly what I'm offering to build for you, not just demo'd it.
- I write production code with logging, error handling, and backup strategies built in from day one - not prototype scripts that fall over in week two.
- Fast iteration loops with verification at each stage. You see working output early and often.
- I'll tell you when AI isn't the right tool. Sometimes a 50-line Python script beats a 500-line agent.
Tell me what you want automated, and we'll scope it together.