Sujith K. Surendran
The model is the easy part. What actually breaks AI projects is the layer underneath it. I focus on wiring agents into enterprise data, identity, security, and governance so they survive production.
- 24+ years in enterprise systems
- 3 production RAG apps shipped
- Fortune-class delivery
- Google Cloud · Claude · OpenAI · MCP
- Hands-on architect
Recent Writing
All writing → From the Tools Index
My 3 highest-conviction picks right now, with a one-line reason each.
- Claude Code skill The fastest path from intent to shipped code I've found.
- LiteLLM tool The cleanest way to route across Claude, GPT-4o, Gemini, and local models behind a single OpenAI-compatible endpoint.
- Context7 MCP mcp Solves the stale-docs problem that plagues AI coding assistants.
Currently
- Designing enterprise agentic systems, RAG applications, and LLM platforms with evaluation, security, and governance built in for production use.
- Mentoring junior engineers and early-career practitioners.
- Volunteering to train school students in technology.
- Writing on Medium, LinkedIn, and Facebook.