Agentic AI Portfolio — AI skill for Claude Code

Sujay Chitnis — Agentic AI Portfolio

**LinkedIn:** https://www.linkedin.com/in/sujaychitnis

This is a snapshot of the agentic AI work I've been building, mostly in Claude Code, alongside a look at what I do day to day in my current enterprise role. It's an overview, not the full picture — happy to walk through any of it in more depth if you want specifics.

Most of what's here is personal, hands-on build work: a multi-agent orchestration framework (Agentic OS) that I designed from scratch, a working job-search system built on top of it that I use daily, an offline document-intelligence system I directed Claude Code to architect end to end, and a few smaller apps in different stages, from submitted-for-review to early build. None of it is dressed up to look further along than it is — a project that's still in alpha says so, a project that's pending approval says so.

Alongside the personal build work, this repo links to my LinkedIn (above), which reflects my current professional role and enterprise background — 23 years in program and delivery leadership, most recently directing AI-driven modernization work at a global scale. That experience is the reason I approach agent-building the way I do: with the same discipline around verification, human sign-off, and accountability that enterprise delivery requires, just applied to a newer set of tools.

I've been using agentic AI tools heavily, not as a novelty but as my actual working method, and I'm building products with them, several of which are live in this portfolio in different stages of maturity. My own direction from here is straightforward: I want to keep growing as a hands-on agentic AI builder, someone who ships real systems with these tools rather than just talking about what they could do.

If something here is interesting, the best next step is a conversation — I'd rather walk you through how a piece actually works than let a one-line description do the talking.