// Full-Stack · AI Engineering · Open Source
Building systems that think — from raw model to production product.
I build AI-powered products that go beyond demos — production-ready systems with reliable inference, thoughtful UX, and clean APIs. My background spans distributed systems, LLM orchestration, and the full web stack.
I believe the most interesting engineering happens when language models meet real constraints: latency budgets, cost ceilings, imperfect data, and actual users. That's where good architecture matters.
Currently based in Pune, MH — open to remote-first teams working on meaningful AI problems.
Languages
AI / ML
Frameworks
Infra
DocMind
Open-source RAG framework for building agentic LLM applications.
DocMind is a Retrieval-Augmented Generation (RAG) application that enables users to upload one or more PDF documents and interact with them through a conversational interface. The system uses semantic search to retrieve the most relevant document chunks and leverages a Large Language Model (LLM) to generate accurate, context-aware answers with source citations.
ResearchAgent
Autonomous multi-step research agent powered by LangGraph and live web search.
A stateful agent that plans, searches, reads, and synthesizes multi-source reports on any topic. Includes a human-in-the-loop approval step and exports to Markdown.
Synthex SaaS
B2B content intelligence platform — AI writes, edits, and scores long-form copy at scale.
Full-stack SaaS with Stripe billing, team workspaces, and a custom fine-tuned GPT-4 variant for brand-voice consistency. Serves 1,200+ paying users.
llm-eval-kit
Lightweight CLI for running structured evals against any OpenAI-compatible endpoint.
Configurable test suites via YAML, supports factuality, toxicity, and custom rubric grading. Integrates with CI pipelines. 800+ GitHub stars.
Data Analyst
Designed and shipped production RAG systems, agentic pipelines, and developer tooling for enterprise clients. Reduced inference costs 40% via prompt compression and caching strategies.
Full-Stack Developer Intern
Built internal ML-assisted fraud detection dashboards and APIs processing 50M+ events/day. Led migration of core services from Ruby to TypeScript.
Whether it's a full-time role, a contract project, or just a question about AI systems — my inbox is always open.