AI/ML Engineer specializing in LLM systems, retrieval-augmented generation (RAG), and production-grade AI pipelines.

github

Pushkin Ranjan  • Peerlist

Charito is a LangGraph-based autonomous engineering runtime that plans, writes, and executes code in real time.It separates planning (architectural reasoning) from execution (file edits, commands, builds) and makes the entire agentic workflow inspectable.Built with:<ol><li>LangGraph (Planner / Architect / Coder nodes)</li><li>Controlled file + shell execution</li><li>Deployment via Render &amp; Vercel</li></ol>Live Link: <a target="_blank" rel="noopener noreferrer" class="break-all link-break text-gray-500 hover:underline" href="https://charito-one.vercel.app/studio">https://charito-one.vercel.app/studio</a> The focus isn’t just code generation — it’s observability, traceability, and reliable execution. I built this to solve the "transparency" problem in AI agents. If you’re a developer working with LangGraph or high-speed Groq inference, I’d love to hear your thoughts on the node-based architecture. Feel free to star the repo, open an issue, or suggest a new "Node" type!