About
AI engineer with hands-on experience building LLM, agentic, and machine-learning systems using Python. Experience spans RAG pipelines, AI infrastructure, backend APIs, model integration, and applied ML across MCP, Hugging Face, FastAPI, FAISS, PyTorch, Gemini, and Azure.
Experience
-
AI Engineer InternTheGrowthVectorMay 2026 – Jun 2026
Engineered scalable AI backends and multi-agent workflows, including a production NLP-to-SQL translation database interface using OpenAI APIs with strict schema validation. Built asynchronous task queues with PostgreSQL-backed persistence, resolved worker queue-lock failures, minimized execution latency, and managed Azure cloud infrastructure, secret configs, and GitHub Actions integration testing.
-
AI InternNikqik Tech. · Remote (Sydney)Jan 2026 – Mar 2026
Designed and deployed an end-to-end RAG architecture for an enterprise meeting intelligence system. Built transcript parsing pipelines, prototyped MCP server and client plugins using the MCP SDK, and developed backend services in Python and JavaScript with MongoDB and Azure.
Education
-
2025
Bachelor of EngineeringJSPM’s Imperial College of Engineering & ResearchComputer Engineering · 2021 – 2025
Skills
Tools / apps / platforms
Projects
-
MeshCardsFastAPI, React, Gemini
AI Learning & Flashcard Generation Platform. Built an AI-powered learning platform transforming documents into structured flashcards and Anki-compatible study material.
-
Zero FrictionLangGraph, Retrieval, Memory
LLM Context Management Infra. Built an LLM context-management layer with priority-based context allocation and integration with retrieval, memory, tools, and LangGraph workflows.
-
InlayOpenAI APIs, SSE, zstd
Middleware & Observability Layer for Coding-Agents. Built transparent middleware for coding-agent traffic with request-level observability and privacy-aware safeguards.
-
Harness-AgentFastAPI, MCP, Next.js, TypeScript
AI Reliability & Simulation Harness. Designed a deterministic software-in-the-loop reliability harness for autonomous agents and built an investigation pipeline from experiment planning through regression verification.