Elliot Shohet.

Full-stack Engineer

I build AI products and developer tools with TypeScript, React, and Node.js—from the interface and APIs to production.

Explore my work
View résumé

Work, from end to end.

BurnBurnOct 2022 — Present
Full-stack Engineer

Health tracking.
AI coaching.
One connected product.

I built Burn’s mobile app and web workspace around shared health records, with streaming AI coaching and consent-based tools for ChatGPT.

  • Mobile and web health workflows on shared TypeScript services
  • Streaming coaching with saved history and retry safeguards
  • Shared sign-in and consent-based ChatGPT tools
  • TypeScript
  • React Native
  • Expo
  • Next.js
  • React
  • PostgreSQL
  • Prisma
  • Bun
  • OpenAI
  • Model Context Protocol (MCP)
  • Better Auth
  • OAuth
  • Google Health
  • Docker Compose
Visit BurnRead case study
Burn mobile dashboard showing nutrition, calories, and meal tracking
Burn’s mobile nutrition dashboard.

Nifty’s

Full-stack Engineer
May 2021 — Oct 2022

At Nifty’s, I engineered smart contracts supporting more than 100,000 NFT mints. The work combined Solidity implementation with TypeScript release tooling and CI/CD across Ethereum and Palm.

  • Solidity
  • TypeScript
  • Ethereum
  • Palm
  • CI/CD
100,000+ NFT mints supported
Read case study

Hypernet Labs

Full-stack Engineer
Feb 2019 — Feb 2020

At Hypernet Labs, I architected layer-two infrastructure for ERC20 payments reaching 25 transactions per second. I also built a modular developer SDK, introduced CI/CD automation, and managed a small engineering team and its technical roadmaps.

  • Ethereum
  • ERC20
  • SDK development
  • Dependency injection
  • CI/CD
25 tx/sec ERC20 payment settlement throughput
Read case study

i2Chain

Full-stack Engineer
Sep 2018 — Feb 2019

At i2Chain, I built Electron, React, and Redux desktop software and Node.js identity-based access controls. I also developed AWS Lambda services and operated a production geth node cluster.

  • Electron
  • React
  • Redux
  • Node.js
  • AWS Lambda
  • Ethereum
Read case study

Skills

Frontend

  • React
  • Next.js
  • CSS
  • Tailwind CSS
  • Redux
  • React Native
  • Expo
  • Electron
  • Three.js
  • React Three Fiber
  • shadcn/ui
  • React Router
  • jQuery

Backend

  • Node.js
  • Express
  • REST APIs
  • WebSockets
  • OAuth
  • Better Auth
  • Bun
  • Zod
  • Sharp
  • Ethereum
  • Web3.js

Data

  • PostgreSQL
  • MongoDB
  • Redis
  • SQL
  • NoSQL
  • Prisma
  • Drizzle ORM
  • SQLite
  • Supabase
  • IPFS

DevOps

  • Docker
  • Docker Compose
  • CI/CD
  • GitHub Actions
  • AWS Lambda
  • Amazon EC2
  • Linux
  • Git
  • Vercel
  • Vercel Sandbox
  • Vercel Workflow
  • Expo EAS
  • AWS Elastic Beanstalk
  • Chef
  • Webpack

Testing

  • End-to-end testing
  • Playwright
  • Jest
  • Vitest
  • Mocha
  • Chai
  • Node.js test runner
  • Bun test
  • Python unittest
  • Truffle
  • Ganache

AI

  • Large language models (LLMs)
  • OpenAI
  • AI SDK
  • Tool calling
  • Model Context Protocol (MCP)
  • AI evaluations
  • Agent harnesses
  • Codex

Languages

  • TypeScript
  • JavaScript
  • Python
  • SQL
  • Solidity
  • Shell scripting

Built to be explored.

Freecoda2026

Your next idea, free

I built an open-source AI assistant across web, desktop, and terminal, with a shared TypeScript agent and reviewable project actions.
  • Shared agent, Electron app, terminal CLI, and Next.js service
  • Separate approvals for file edits and commands
  • Conversation sync and PostgreSQL usage accounting
  • TypeScript
  • React
  • Next.js
  • Electron
  • Bun
  • PostgreSQL
  • AI SDK
  • Model Context Protocol (MCP)
  • Better Auth
  • Drizzle ORM
  • Zod
  • Tailwind CSS
Visit FreecodaGitHubRead case study
Oct 2026

Pets

A local Next.js demo for pet profiles, medical records, medications, and care summaries, backed by Prisma and PostgreSQL.

  • TypeScript
  • React
  • Next.js
  • Prisma
  • PostgreSQL
  • Zod
  • Sharp
  • Docker Compose
Source code
2015

Machine Learning & Doodle Jump

Contributed to a UC Davis team project applying Q-learning to HTML5 Doodle Jump, with training controls, performance charts, and browser persistence.

  • Q-learning
  • HTML5
  • JavaScript
Source code