Anirudh's Website </>

Resume·He/Him·Updated October 2026

Anirudh Vasudevan

Full-stack and AI engineer in San Francisco. Member of Technical Staff at Rivo.

anirudhvasudevan11@gmail.comWebsiteLinkedIn ↗GitHub ↗

Download PDF 151 KB Open the PDF ↗

Education

University of Minnesota, Twin Cities

MN, USA

Master of Science in Computer Science GPA 3.89/4

2023 to 2025

SRM Institute of Science and Technology

TN, India

Bachelor of Technology in Computer Science and Engineering GPA 9.2/10

2019 to 2023

IIT Madras

TN, India

Diploma in Data Science

2021 to 2023

Skills

Languages and frameworks
  • JavaScript
  • TypeScript
  • Python
  • React
  • React Native
  • Next.js
  • Flask
Database and backend
  • SQL
  • PostgreSQL
  • Supabase
  • Firebase
  • RESTful APIs
  • Pinecone
  • Microservices Architecture
AI and machine learning
  • LangChain
  • LangGraph
  • RAG
  • Agents
  • Fine-tuning
  • MCP
  • Ollama
  • LiveKit
  • Hugging Face
Analytics and growth
  • Google Tag Manager
  • GA4
  • Mixpanel
  • Statsig
  • Meta Ads
  • TikTok Ads
  • Reddit Ads
  • Google Ads
Cloud and DevOps
  • AWS
  • Docker
  • Git
  • GitHub
  • Figma
  • Agile
  • System Design
  • CI/CD
  • Playwright
  • Maestro
  • Jest

Professional experience

Rivo

San Francisco, CA, US

Member of Technical Staff

Dec 2025 to Present

  • Deployed production conversational and agentic AI in Python on cloud microservices, plus LiveKit voice agents that capture leads and an internal support chatbot.
  • Implemented TanStack Query caching with endpoint-level invalidation, cutting homepage load time 87% and improving p95 across critical flows.
  • Own and maintain the iOS and web apps on one React Native codebase migrated from Next.js, hardened with Playwright and Maestro E2E tests and Sentry monitoring with testing workflows.
  • Own product analytics and growth for a B2C fintech app, running full-funnel tracking and paid ads across TikTok, Meta, Reddit, and Google.
  • Shipped a scalable, admin-configurable referral system the team extends without engineering.
  • Built an automated pipeline that generates UGC-style video ads end to end.

Nonlinear

San Francisco, CA, US

Founding Engineer

Aug 2025 to Dec 2025

  • Engineered the Universal Node System for AI workflows with dynamic property handling and a Zustand-powered state caching, reducing configuration time by 45% and enabling faster feature rollout across the platform.
  • Built Agentic RAG pipelines powering context-aware AI agents and a Teams bot, reducing retrieval latency by 60% and tripling user engagement post-launch.

Project experience

RUA

Privacy-First AI Assistant for Caregivers

  • Solo-built and shipped an iOS (TestFlight) and web app for family caregivers, with an integrated voice agent for the people in their care and a privacy-first design that keeps records encrypted and grounded to the user's own data.
  • Built the paperwork automation as configurable node-and-edge workflows that draft insurance appeals, prior-auth packets, and Medicaid renewals, cutting a denial appeal from 3.5 hours to 12 minutes.
  • Shipped the core features: medication and inventory tracking with refill runway, ER sheets, appointment prep, and Ask RUA, the AI assistant that answers from the user's records.

YouTube Shorts Factory

Automated UGC Video Engine and Dashboard

  • Built an end-to-end pipeline that turns a topic into a published Short (LLM scripting, Fal.ai image and video, ElevenLabs TTS, Whisper subtitles), fully auto-uploaded.
  • Runs 4 channels on daily cron via GitHub Actions, with a review-and-approve dashboard.

Patents, publications, and achievements