David Zhang

An AI-powered research assistant that performs iterative, deep research on any topic by combining search engines, web scraping, and large language models. The goal of this repo is to provide the simplest implementation of a deep research agent - e.g. an agent that can refine its research direction overtime and deep dive into a topic.

Get structured, fully typed, and validated JSON outputs from OpenAI and Anthropic models.

Run Claude Agent (Claude Code) in a sandbox, control it via websocket

LLM powered retrieval engine designed to process a ton of sources to collect a comprehensive list of entities.

Fully typed & consistent chat APIs for OpenAI, Anthropic, Groq, and Azure's chat models for browser, edge, and node environments.

The Typescript-first prompt engineering toolkit for working with chat based LLMs.

A TypeScript library providing OAuth2 authentication utilities for Model Context Protocol (MCP) clients. This library simplifies the process of adding OAuth authentication to MCP client implementations.

SocketStream iOS client

Simple API for creating a webhook endpoint to capture webhook requests. Built for AI agents that cannot receive pushed webhooks, this allows agents to pull webhook events as needed.

For when I need to start a llm app from scratch

My VIM config. Clone to ~/.vim & copy .vimrc to ~/.vimrc

IE Breakup Page

Just testing out the capabilities of gmail extensions

Personal website

Roadtrip iOS App