I'm a dual-degree student in the UC Berkeley MET program studying Electrical Engineering & Computer Science + Business. I'm passionate about building tools at the intersection of AI, systems, and user experience — especially ones that solve real problems, run efficiently, and scale with purpose.
- 🔬 Revyl (YC W24): Interning at a Y Combinator-backed startup where I built a Model Context Protocol (MCP) server to orchestrate mobile E2E tests, reduced p95 latency by 40%, and deployed a Slackbot for real-time observability data retrieval.
- 🗣 CodeVox: A mobile voice-first coding assistant using Whisper + Claude Code that generates patches from spoken commands; scaled to 150+ devs in 3 months.
- 🧠 LDT Compliance Copilot: An AI agent that uses NVIDIA NeMo + LangChain to parse FDA/CLIA regulations and generate compliance gap reports in seconds.
- 🍽 NutriGenie: Built an AI meal-planning agent that aligns macros, budgets, and fitness goals using grocery store APIs and OpenAI — scaled to 300+ users in 2 months.
- 📚 AmigoAI: Created an adaptive SAT practice-test module and improved tutoring accuracy by 18% using retrieval-augmented generation.
- 💪 Uberlast: Co-founded a personalized fitness app with pose estimation and progressive overload tracking — reached 25%+ 30-day retention.
- 🔔 Revyl Slackbot: Built and deployed an NLP-to-SQL Slackbot to surface live service diagnostics, replacing a Grafana workflow and cutting triage cycles.
Python TypeScript React Swift FastAPI LangChain Anthropic Claude
OpenAI API PostgreSQL Redis Docker AWS Lambda / EC2 Firebase NVIDIA NeMo
- ✉️ Email: sidv@berkeley.edu