framework
ClosedClaw
Local capsule-based agent orchestration with LangGraph routing, sqlite-vec search, approval gates, and JSONL audit trails.
AI/ML engineer building agent infrastructure
LangGraph workflows, retrieval and evaluation pipelines, browser agents, automation runtimes, and internal copilot systems designed to be observable, usable, and hard to misread.
01 / Live Systems
Built around explicit loops: plan, retrieve, act, observe, evaluate, and recover.
framework
Local capsule-based agent orchestration with LangGraph routing, sqlite-vec search, approval gates, and JSONL audit trails.
runtime
Zero-dependency workflow runtime for YAML-defined agent, function, and tool steps with SQLite checkpoints and replay.
browser agent
DOM-driven browser agent using Gemini native function calls, approval modes, Playwright execution, and run evidence logs.
digital twin
LangGraph water-management assistant that classifies prompts, extracts simulation parameters, and analyzes stable ranges.
02 / Experience
Internships and production-adjacent work that shaped how I think about agents, retrieval, and evaluation.
LangGraph workflows in Siemens Capital Copilot and an agentic productivity framework for developer workflows.
May - Jul 2025 Software Engineering Intern, DocLens.aiMedical necessity evaluation, hybrid retrieval, RAG pipelines, RAGAs, and DeepEval for insurance workflows.
May - Jul 2024 AI Engineering Intern, Bicycle AIContext-aware SQL generation and benchmarking frameworks for LLM output quality.
03 / Writing
Longer explanations for agent systems, structured content, and interface decisions.
04 / Background
Enough context for recruiters without letting credentials overpower shipped systems.
Education
B.Tech in Computer Science Engineering, 2022-2026. CGPA 9.42/10.0.
Certifications
DeepLearning.AI supervised ML, Hugging Face agents and LLM courses, Google Cloud LLM/GenAI/Transformers.
Leadership
President, Vice President of Education, Program Quality Director, and contest wins across speaking formats.
05 / Contact
Email is best for direct conversations. GitHub and writing are the clearest places to inspect the work.