portfolio / AI Engineer
AI Engineer focused on building production-ready AI systems across Voice AI, RAG, agentic workflows, memory infrastructure, evaluations, and LLM-powered applications.
I enjoy combining AI with backend engineering to build reliable retrieval pipelines, orchestration systems, observability, and scalable APIs, with a strong interest in improving latency, retrieval quality, and overall system reliability.
Growth Engineer – Super 30
• Worked on post-purchase funnel optimization, focusing on learner activation, engagement, cohort retention, and reducing course drop-offs. • Contributed to improving course completion from 5% to 20% (4x lift) over 2–3 months, a 15 percentage-point increase through learner engagement and retention initiatives
Co-Founder
Co-founded an ad-sponsored bottled water venture inspired by FreeWater, developing a two-sided business model where advertisers fund free drinking water distribution through branded packaging. • Evaluated unit economics and commercial viability, analyzing manufacturing, logistics, advertiser pricing, and minimum order quantities; identified thin contribution margins and advertiser acquisition challenges
Software Developer -Intern
Led a team of interns in building a RAG pipeline for searching companies’ internal documents. Designed and optimized the document ingestion and retrieval workflow, reducing retrieval latency from approximately 4–5 seconds to around 200 milliseconds while improving access to relevant internal knowledge.
Full Stack Developer -Intern
Contributed to the development of an EdTech platform using TypeScript and React. Built reusable user-interface components, integrated frontend features with backend APIs, resolved functional and UI issues, and collaborated with the development team to deliver and improve product features.
Built a real-time conversational voice agent using LiveKit Agents, streaming STT, VAD, LLMs, and TTS; implemented interruption handling, acoustic/textual state classification, and state-aware response strategies. • Reduced measured voice-response latency from 7–8 s to 3–3.5 s by optimizing turn detection and classifier initialization; built per-turn observability for latency attribution and asynchronous event ownership.
Built a reusable AI memory infrastructure layer that enables LLM applications to extract, store, update, and retrieve long-term user and conversation knowledge. The system supports semantic and episodic memories, deterministic deduplication and supersession, hybrid lexical + vector retrieval, importance/recency-aware ranking, conversation summaries, and per-user memory isolation. Designed the memory lifecycle from LLM-based extraction through policy-driven validation and persistence, with FAISS-based semantic search and evaluation tooling to measure retrieval quality across memory benchmarks.
Built an enterprise agentic AI platform that combines LLM reasoning with deterministic validation for reliable workflow automation. Implemented tool orchestration, DAG execution, evidence-grounded reasoning, RAG, evals, observability, provenance, execution budgets, and governed actions with human approval.
Built a perpetual futures exchange backend with a deterministic matching engine supporting limit/market orders, leveraged long/short positions, margin management, fills, liquidation-price calculation, and real-time order-book updates. Designed an event-driven architecture using Redis Streams, REST APIs, WebSockets, and Binance market-price feeds.