# Bun Projects at AI Tinkerers

> Canonical HTML: https://aitinkerers.org/technologies/bun
> Markdown URL: https://aitinkerers.org/technologies/bun.md
> Technology record last updated: 2026-09-18T15:14:00Z
> Generated: 2026-09-22T01:43:56Z

Bun is the fast, all-in-one JavaScript runtime and toolkit: it replaces Node.js, npm, and multiple development tools with a single, cohesive executable.

Bun delivers a high-performance JavaScript runtime, built from scratch in Zig and powered by Apple's JavaScriptCore engine (not V8). It functions as a drop-in replacement for Node.js, but unifies the entire development workflow: the runtime, package manager (e.g., `bun install`), bundler (`bun build`), and test runner (`bun test`) are all integrated. This architecture dramatically reduces startup times and memory usage, offering significant speed gains, especially in tasks like package installation (up to 30x faster than npm) and server-side rendering.

- Official technology site: https://bun.sh/
- Public AI Tinkerers demos and talks: 19
- Result page: 1 of 1

## Recent Public Talks and Demos

### [Phat-deck - using strict schemas as the ultimate Human-LLM contract](https://wellington.aitinkerers.org/talks/rsvp_u819Qxav0Pg)

I built phat-deck, a pure CLI config-as-code slide compiler that transforms raw YAML into pixel-perfect PDF, PNG, and LinkedIn/Reddit carousels without a frontend GUI (yet). I wrote the entire codebase over a single weekend exclusively through the Gemini chat window. For the demo, I will open the terminal, edit the YAML configuration live, and show the exact workflow of how I use strict JSON schemas to define the unbreakable contract between human intent, the LLM's output, and the rendering engine. I'll compile a multi-page, high-res deck directly from the CLI in real-time.

- Event context: AI Tinkerers Wellington: Inaugural Meetup · 26 August 2026 — 2026-08-26 — Wellington
- Public talk page: https://wellington.aitinkerers.org/talks/rsvp_u819Qxav0Pg

### [Does your coding harness actually do anything? Measuring it, fairly.](https://boston.aitinkerers.org/talks/rsvp_bx0CiDi1_uo)

Everyone's bolting "harnesses" onto coding agents — Superpowers, GSD, Agent-Skills, Compound Engineering — but does the scaffolding actually make the model build better software, or just feel better? CodingHarness.xyz is an open eval that pits these frameworks head-to-head: same spec, same model, isolated sandboxes, and an evidence-based rubric (does it meet the PRD? is the code any good?). I'll share the surprising result — an inverse-scaling effect lifted straight from a recent Xiaomi paper and reproduced on real runs: harnesses help most exactly where the base model is weakest, and barely at all where it's already strong. Even better, it's framework-dependent — one framework tracks the curve almost perfectly, another inverts it and only helps strong models. Live demo of the eval studio: watch two frameworks build the same app, stream the agent's work turn-by-turn, and see the marginal-gain scorecard light up green/red. Plus a sneak peek at turning framework competitions into a spectator sport (think FIFA bracket, but for coding agents).

- Event context: AI Tinkerers Boston: GTM Agentic AI Launch — 2026-06-29 — Boston
- Public talk page: https://boston.aitinkerers.org/talks/rsvp_bx0CiDi1_uo

### [GTM in the Terminal — Agent-driven personalized outbound](https://sf.aitinkerers.org/talks/rsvp_O2H7CQUJ6IU)

Cyrus is the cofounder of The Synthesis Company (YC S24). Previously, he was a core contributor to Stanford NLP's DSPy framework and cofounder of WOMBO.ai (100M+ downloads). He'll be demoing GTM-OS, a CLI and TUI he used to programmatically drive hyper-personalized LinkedIn outbound. For AI tinkerers he built usesocial.dev, a CLI to connect your agents to LinkedIn / X and build your own workflows and systems.

- Event context: AI Tinkerers San Francisco: GTM Engineering Track — 2026-05-19 — San Francisco
- Public talk page: https://sf.aitinkerers.org/talks/rsvp_O2H7CQUJ6IU

### [Hive: Local-First AI Gateway](https://bogota.aitinkerers.org/talks/rsvp_ukV86hfl3Q0)

Explore building a local-first AI agent gateway with Bun and SQLite. Learn to orchestrate autonomous agents across multiple communication channels, manage tools, and switch LLM providers seamlessly.

- Event context: AI Tinkerers Bogotá: Demos for builders — 2026-04-23 — Bogotá
- Public talk page: https://bogota.aitinkerers.org/talks/rsvp_ukV86hfl3Q0

### [Building a Real-Time Terminal Messenger on LinkedIn's Private APIs with Bun, Ink, and a Git-Backed Message Store](https://seattle.aitinkerers.org/talks/rsvp_0AqXN-gxTTM)

Allman is a two-layer system for LinkedIn messaging from the terminal: a CLI that syncs, sends, and streams LinkedIn messages into a git-versioned file store, and a TUI that renders it as a full two-pane Ink/React terminal messenger with live updates, auto-backfill, and real-time presence. The demo shows the TUI in action — navigating conversations, composing replies, watching the status bar update as messages stream in over LinkedIn's SSE channel, and kicking off a sync that backfills an entire conversation history while showing live progress counts. Under the hood, the TUI never touches the network directly. Every write — sends, syncs, searches — shells out to the lilac binary, which is embedded inside the compiled TUI executable as a bundled asset. Every read — conversation list, message thread, slug lookup — goes straight to JSONL files on disk, so navigation is instant with zero subprocess overhead per keystroke. The key learning or takeaway is that private APIs, even when obfuscated, are easily reverse-engineered by coding harnesses. Claude Opus 4.6 processed the compiled LinkedIn binary overnight, completely autonomously.

- Event context: AI Tinkerers Seattle: GTM Engineering — April Meetup — 2026-04-23 — Seattle
- Public talk page: https://seattle.aitinkerers.org/talks/rsvp_0AqXN-gxTTM

### [From AI Slop to Beautiful: Automated Frontend Redesign with TinyFish + Claude](https://ho-chi-minh-city.aitinkerers.org/talks/rsvp_JSBl9hedRKQ)

A pipeline that takes any ugly AI-generated frontend, researches what best-in-class websites in your industry actually look like using TinyFish’s deep web research, generates a design brief from those findings, and then dispatches a Claude agent to rewrite the UI — all ending with a GitHub PR you can review and merge. One command turns a generic AI-generated site into something that looks like a real product.

- Event context: AI Tinkerers Ho Chi Minh City: AI Coding Agents & Orchestrators — 2026-04-18 — Ho Chi Minh City
- Public talk page: https://ho-chi-minh-city.aitinkerers.org/talks/rsvp_JSBl9hedRKQ

### [Ingeniería Inversa de Plugins de Claude Code: De Shanghai a Medellín](https://medellin.aitinkerers.org/talks/rsvp_fSouJaWMtIE)

Voy a mostrar cómo construí notebooklm-ai-plugin estudiando plugins del Shanghai AI Laboratory y herramientas de investigación científica. El patrón que encontré: scripts que usan cookies del navegador para comportarse como el sitio web de Google y hacer llamadas RPC directas, aprovechando la capa gratuita de estas plataformas sin necesidad de API keys. La charla cubre cómo reverse engineeré ese patrón, cómo lo adapté con TypeScript y Bun para construir un plugin con 9 tipos de artifacts, y cómo el manifest .claude-plugin lo hace distribuible con un solo comando. Sin diapositivas. Solo terminal. El repo abierto todo el tiempo.​​​​​​​​​​​​​​​​

- Event context: AI Tinkerers Medellín #20 - SalesAds - 16 de Abril — 2026-04-16 — Medellín
- Public talk page: https://medellin.aitinkerers.org/talks/rsvp_fSouJaWMtIE

### [Anita: A Proactive Synthetic Organism](https://cologne.aitinkerers.org/talks/rsvp_10X8PdiOc28)

Anita is a proof-of-concept architecture for building proactive AI agents that perceive and act on their own, instead of waiting for a prompt. Inspired by the Synths from the TV series "Humans" and by the OpenClaw project, the architecture is modeled after biological organisms: receptors perceive continuous streams of stimuli and react to them. The agent has "organs", "senses", a central "nervous system", and declarative "reactions". The key goal is extensibility, not just by the developer, but by Anita herself. In the demo, I'll fire stimuli into the running system and show how perceptions flow through reactions and produce actions.

- Event context: AI Tinkerers Cologne 4: Live Technical Demos — 2026-04-16 — Cologne
- Public talk page: https://cologne.aitinkerers.org/talks/rsvp_10X8PdiOc28

### [What did we learn from the March 31st Claude Code Source Leak?](https://seattle.aitinkerers.org/talks/rsvp_P82HqL1GQaM)

This technical analysis, presented by Daniel Motles of Qumulo in April 2026, dissects the architecture and leaked details of Claude Code. Claude Code is a Bun-native TypeScript application with a React and Ink terminal UI, functioning as a productized agent operating system. The system is composed of 1,902 source files totaling 512,685 lines of code. Its technical stack includes the Anthropic SDK, MCP integrations, and native tools for Git, Shell, and LSP. A major finding is the discovery of KAIROS, an always-on autonomous daemon that runs heartbeat prompts every 30 seconds to identify tasks, fix errors, and manage memory consolidation. Other unreleased features include Ultraplan for deep thinking via remote Opus variants, Coordinator Mode for spawning sub-agent forks, and an internal Undercover Mode that strips AI signals from public commit histories. The system utilizes a heavyweight persona driven by a massive system prompt and a four-tier MD memory system covering users, feedback, projects, and references. To preserve context window space, Claude Code uses deferred tooling and a local LRU cache to prevent redundant token consumption. Local search is handled by a parallelized Rust-based engine called RipGrep (rg). The analysis concludes that the engineering "plumbing"—the tools, memory, and safeguards—constitutes the product's primary competitive moat. While the leak has spurred open-source clones, it has also raised enterprise concerns regarding release hygiene and intellectual property safety.

- Event context: AI Dev Tool Track — Seattle Meetup - April 13th, 2026 — 2026-04-14 — Seattle
- Public talk page: https://seattle.aitinkerers.org/talks/rsvp_P82HqL1GQaM

### [Hive — Building a Local-First AI Agent Gateway with Bun, SQLite & Multi-Channel Orchestration](https://pereira.aitinkerers.org/talks/rsvp_xTm-4SxgHLw)

Hive is an open-source, local-first AI agent gateway built entirely with Bun and TypeScript. It lets you deploy autonomous AI agents that communicate across Telegram, Discord, WhatsApp, and CLI — all orchestrated through a single SQLite database with zero cloud dependency. The architecture follows a native Context Engine + ACE (Agentic Context Engineering) pattern: agents are stored as data, not hardcoded classes, and a Coordinator agent dynamically spawns sub-agents, assigns tools, and routes tasks. Hive supports multiple LLM providers (OpenAI, Gemini, Ollama, OpenRouter, Moonshot) through a unified provider adapter layer, making it truly model-agnostic. Built independently from Bogotá, Colombia and published on npm as @johpaz/hive.

- Event context: AI Tinkerers Pereira — The Builders Lab — 2026-03-25 — Pereira
- Public talk page: https://pereira.aitinkerers.org/talks/rsvp_xTm-4SxgHLw

### [AI managed Meta ads](https://seattle.aitinkerers.org/talks/rsvp_xGxD9rmrok8)

AI agent to generate assets and manage meta ads

- Event context: AI Tinkerers Seattle: GTM Engineering Kickoff — Building AI for Growth — 2026-02-26 — Seattle
- Public talk page: https://seattle.aitinkerers.org/talks/rsvp_xGxD9rmrok8

### [Collaborator: Lobster-powered critical thinkig partner](https://nyc.aitinkerers.org/talks/rsvp_Lh8ujzeqlDQ)

Collaborator is a set of open-source tools that turn an OpenClaw agent into a thinking partner. Code walkthrough and live demo of three building blocks: (1) Seedvault — markdown sync between humans and agents, giving the agent a living knowledge base. (2) Reflection pipelines — chained sub-agents that run during OpenClaw's heartbeat, digesting notes into goals, blockers, and recommendations. (3) Collab MCP — a TypeScript server that packages the agent's persona and accumulated knowledge into a portable identity, invokable inside Claude Code or Codex.

- Event context: 🦞Demo Night: OpenClaw ft Convex — 2026-02-17 — New York City
- Public talk page: https://nyc.aitinkerers.org/talks/rsvp_Lh8ujzeqlDQ

### [Verifying vibe code](https://seattle.aitinkerers.org/talks/rsvp_H9IOQoXp-44)

The barrhawk premium e2e mcp for claude cli allows you to take "trust me bro" "lgtm" vibe code and properly test it in a end to end, holistic thinking process

- Event context: AI Tinkerers Seattle: January Meetup — 2026-01-31 — Seattle
- Public talk page: https://seattle.aitinkerers.org/talks/rsvp_H9IOQoXp-44

### [How I hacked the hottest SF startup](https://poland.aitinkerers.org/talks/rsvp_LATkC2SrA1A)

The story of how I leaked Poke's system prompt and reverse engineered its architecture to know what makes it so good

- Event context: AI Tinkerers Poland #6 - Meetup in Warsaw (26th November) — 2025-11-26 — Poland
- Public talk page: https://poland.aitinkerers.org/talks/rsvp_LATkC2SrA1A

### [Torque: Like React, but for Synthetic Datasets](https://sf.aitinkerers.org/talks/rsvp_ZSKhh9gzUN8)

I'll show how with Torque in a matter of minutes you can create a high quality non trivial dataset for building and testing your initial evals: - Building a basic Q&A dataset - Adding tool-calling patterns - Modeling async operations (web searches, long-running tasks) - Scaling from 10 to 1000 examples with one parameter change

- Event context: Evals are Over - Evals are Back - AI in Prod @ Okta — 2025-10-31 — San Francisco
- Public talk page: https://sf.aitinkerers.org/talks/rsvp_ZSKhh9gzUN8

### [Agentify Agent: Converting Pre-Built Agents into the Agentuity Framework](https://miami.aitinkerers.org/talks/rsvp_uJjSEJ3mvzw)

The Agentify Agent is an agent built on Agentuity that takes links to open-source AI agent examples (like CrewAI or Mastra) and automatically converts them into runnable Agentuity agents. It analyzes docs, extracts runnable TypeScript snippets, sets up the correct folder structure, and opens a pull request with a ready-to-run example. I’ll walk through how the agent interacts with Devin (the AI software engineer), how the prompt architecture was designed, and what we learned about agent onboarding.

- Event context: AI Tinkerers Miami Meetup — Fall Demo Night at The Lab — 2025-10-28 — Miami
- Public talk page: https://miami.aitinkerers.org/talks/rsvp_uJjSEJ3mvzw

### [Ona: Building AI Agent Workflows](https://boston.aitinkerers.org/talks/rsvp_9vqE3e0W38Y)

I'll show how we built Ona, a web app where AI agents autonomously execute complex workflows across cloud services.

- Event context: AI Tinkerers Boston Meetup at Publicis Sapient — 2025-09-29 — Boston
- Public talk page: https://boston.aitinkerers.org/talks/rsvp_9vqE3e0W38Y

### [Fixing weight loss with AI](https://poland.aitinkerers.org/talks/rsvp_PaopIPqzHoQ)

We're building slimtwin: a platform that uses your data - from smart scales, wearables, and bloodwork - to build a digital twin of your health. A system that adapts in real time. That learns with you. That reflects your body today - not last month. I'll be presenting an early version of our product targeted at dietitians.

- Event context: AI Tinkerers Poland #4 - Meetup in Warsaw (June) — 2025-06-26 — Poland
- Public talk page: https://poland.aitinkerers.org/talks/rsvp_PaopIPqzHoQ

### [How to make an MCP server](https://miami.aitinkerers.org/talks/rsvp_Lz9P-BgdAvo)

In this short talk I'll go over what is MCP and how someone can ship their own MCP. As an example I'll be showing and demoing an MCP server I have shipped that has users called near-mcp

- Event context: AI Tinkerers - Miami: May Meetup at The LAB — 2025-05-29 — Miami
- Public talk page: https://miami.aitinkerers.org/talks/rsvp_Lz9P-BgdAvo

## Related Technologies

- [TypeScript](https://aitinkerers.org/technologies/typescript) ([Markdown](https://aitinkerers.org/technologies/typescript.md)) — 205 public demos
- [OpenAI API](https://aitinkerers.org/technologies/openai-api) ([Markdown](https://aitinkerers.org/technologies/openai-api.md)) — 520 public demos
- [Claude Code](https://aitinkerers.org/technologies/claude-code) ([Markdown](https://aitinkerers.org/technologies/claude-code.md)) — 215 public demos
- [React](https://aitinkerers.org/technologies/react) ([Markdown](https://aitinkerers.org/technologies/react.md)) — 220 public demos
- [Anthropic API](https://aitinkerers.org/technologies/anthropic-api) ([Markdown](https://aitinkerers.org/technologies/anthropic-api.md)) — 66 public demos
- [Codex](https://aitinkerers.org/technologies/codex) ([Markdown](https://aitinkerers.org/technologies/codex.md)) — 44 public demos
- [Gemini](https://aitinkerers.org/technologies/gemini) ([Markdown](https://aitinkerers.org/technologies/gemini.md)) — 188 public demos
- [Claude](https://aitinkerers.org/technologies/claude) ([Markdown](https://aitinkerers.org/technologies/claude.md)) — 174 public demos
- [CLI](https://aitinkerers.org/technologies/cli) ([Markdown](https://aitinkerers.org/technologies/cli.md)) — 9 public demos
- [Commander](https://aitinkerers.org/technologies/commander) ([Markdown](https://aitinkerers.org/technologies/commander.md)) — 4 public demos
- [MCP](https://aitinkerers.org/technologies/mcp) ([Markdown](https://aitinkerers.org/technologies/mcp.md)) — 129 public demos
- [OpenClaw](https://aitinkerers.org/technologies/openclaw) ([Markdown](https://aitinkerers.org/technologies/openclaw.md)) — 50 public demos
- [SQLite](https://aitinkerers.org/technologies/sqlite) ([Markdown](https://aitinkerers.org/technologies/sqlite.md)) — 28 public demos
- [Supabase](https://aitinkerers.org/technologies/supabase) ([Markdown](https://aitinkerers.org/technologies/supabase.md)) — 90 public demos
- [Agentuity SDK](https://aitinkerers.org/technologies/agentuity-sdk) ([Markdown](https://aitinkerers.org/technologies/agentuity-sdk.md)) — 1 public demo
- [Amazon Bedrock](https://aitinkerers.org/technologies/amazon-bedrock) ([Markdown](https://aitinkerers.org/technologies/amazon-bedrock.md)) — 20 public demos
- [@anthropic-ai/sdk](https://aitinkerers.org/technologies/anthropic-ai-sdk) ([Markdown](https://aitinkerers.org/technologies/anthropic-ai-sdk.md)) — 2 public demos
- [Browser cookie authentication](https://aitinkerers.org/technologies/browser-cookie-authentication) ([Markdown](https://aitinkerers.org/technologies/browser-cookie-authentication.md)) — 1 public demo