I build production-grade AI agents and MVPs that go beyond demos — fully deployed, observable, and built to scale. Whether you need a multi-agent orchestration system, an RAG pipeline, an LLM-powered workflow, or a full-stack AI MVP shipped fast, I bring both the AI engineering depth and the product instinct to get it done right. What I build: Multi-agent systems with LangGraph (parallel agents, human-in-the-loop, checkpointing) RAG pipelines with vector search, chunking strategies, and eval harnesses (RAGAS, Langfuse) Agentic workflows with Temporal for long-running, fault-tolerant processes Full-stack AI MVPs — FastAPI backend + Next.js frontend, Stripe billing, auth, deployed LLM integrations (Claude, OpenAI, Gemini) with structured outputs and tool use Observability setup — Langfuse tracing, RAGAS evals, production monitoring Why work with me: I've shipped three live AI products independently — a multi-agent media intelligence platform (LangGraph + Temporal + RAGAS), an AI image generation SaaS (LoRA fine-tuning on Flux Pro), and a decentralized monitoring network (Solana, recipient of a $30K foundation grant). I move fast, communicate clearly, and deliver systems you can actually maintain. Deliverables include: source code, deployment, documentation, and a walkthrough session.
Yaswanth Adimulam
AI engineer with 3+ years of software engineering experience, holding a B.Tech in Electrical Engineering from IIT Kharagpur, one of India’s top engineering institutions, class of 2022. After three years at Enphase Energy as an iOS engineer, left to start PingProof Labs and build AI products full time, with the past several months focused on production AI systems. Shipped three live products: MediaSense AI, a multimodal brand intelligence platform with LangGraph agents and Temporal workflows.
Posted Jul 17, 2026
