# Beads Projects at AI Tinkerers

> Canonical HTML: https://aitinkerers.org/technologies/beads
> Markdown URL: https://aitinkerers.org/technologies/beads.md
> Technology record last updated: 2026-02-22T20:13:12Z
> Generated: 2026-08-24T21:39:49Z

Naturbeads delivers high-performance, cellulose-based microspheres, directly replacing non-biodegradable plastic microparticles across industrial and consumer sectors.

This is Naturbeads: a critical technology tackling global microplastic pollution at the source. We engineer patented, 100% biodegradable cellulose microspheres to match the performance of plastic microparticles in demanding applications. Our innovative particle technology ensures precise control over size and porosity, offering a sustainable drop-in replacement for sectors like cosmetics, life science (e.g., biocompatible carriers), and industrial coatings. We eliminate the environmental footprint while maintaining the smooth texture and stable performance clients expect, ensuring a seamless transition away from persistent plastic ingredients.

- Official technology site: https://naturbeads.com
- Public AI Tinkerers demos and talks: 6
- Result page: 1 of 1

## Recent Public Talks and Demos

### [/spawn-team: A Self-Healing Multi-Agent Coding Pipeline with Claude Code + Beads](https://seattle.aitinkerers.org/talks/rsvp_yI3yt6tcSig)

A workflow that decomposes a software feature into parallel implementation arms, farms each arm to a Codex agent running in an isolated git worktree, enforces completion ordering through a persistent DAG, auto-retries failed arms with exact judge findings, and merges clean — all resumable across session breaks. The demo shows this running live on a real codebase: pouring the DAG template, watching the explorer dynamically bond implementation arms after reading the code, two Codex agents working in parallel worktrees, the judge failing and retrying, then bd graph showing the full live state.

- Event context: AI Dev Tools Track - Seattle — 2026-05-06 — Seattle
- Public talk page: https://seattle.aitinkerers.org/talks/rsvp_yI3yt6tcSig

### [An Open-Source Harness for Long-Running Autonomous Agentic Development in Production](https://lausanne.aitinkerers.org/talks/rsvp_2noswA3klrA)

I built an open-source agentic coding CLI plugin that packages three requirements for reliable autonomous coding into one system: persistent semantic memory, mechanical feedback loops, and navigable task structure. It runs multi-session agentic work in dependency order, each task through a full cycle of specification, planning, test-driven implementation, multi-model review, and lesson extraction, with no shared context between sessions. I'll demo it through its benchmark: a fully functional browser-based DAW (subtractive synth, drum machine, piano roll, mixer, effects chain) built in 20 hours of unattended autonomous work. This is the same challenge Anthropic ran internally with their unpublished harness (https://www.anthropic.com/engineering/harness-design-long-running-apps, March 24 2026). The harness is open-source, and I'm deploying features and apps in production at my tech company.

- Event context: AI Tinkerers Lausanne April 2026 Meetup — 2026-04-30 — Lausanne
- Public talk page: https://lausanne.aitinkerers.org/talks/rsvp_2noswA3klrA

### [Developing large features without overloading the model context](https://poland.aitinkerers.org/talks/rsvp_aa_j6rU5fzc)

Using divide and conquer strategy to not overload the LLM context when working on large features. It can be easily incorporated into any coding agent thanks to the beads project. I'll show how to integrate that into your workflow and use it to improve the quality of output when developing large features.

- Event context: AI Tinkerers Poland - Meetup in Gdańsk #1 — 2026-04-23 — Poland
- Public talk page: https://poland.aitinkerers.org/talks/rsvp_aa_j6rU5fzc

### [Metadata Driven AI Pipelines](https://seattle.aitinkerers.org/talks/rsvp_3eYiO5c3fGw)

Long running tasks are always a balance of control and agency. As we slowly move away from the CLI to task managers (Jira, Linear, Beads, etc) we will need to figure out ways to find a good balance between control and agency. If control is pure code and agency is purely agentic, then AI pipelines look to be something in between. Code that executes and validates LLM output, but relies on the intelligence of the model to do a majority of the work. I will show how I am designing the AI pipelines and using metadata from a task manager (Linear in this case) to control which pipeline executes for different types of tasks (bugs, features, etc)

- Event context: AI Tinkerers Seattle Meetup: Dev Tools Track — March 9th, 2026 — 2026-03-10 — Seattle
- Public talk page: https://seattle.aitinkerers.org/talks/rsvp_3eYiO5c3fGw

### [Bead Manager](https://boston.aitinkerers.org/talks/rsvp_CMfZIqIUI5k)

Bead Manager is a proof-of-concept web-based user interface for Steve Yegge's Beads project and GasTown agent orchestration system. Bead Managers goals are to: 1. Provide non-technical stakeholders visibility into work-in-progress 2. Provide all users better high-level visibility across the the work 3. Enable stakeholders to provide strategic direction through the web interface 4. Surface and visualize Beads concepts like molecules, bonds, protos, and formulas

- Event context: {Cancelled due to Blizzard Warning } AI Tinkerers Boston: February 2026 Meetup — 2026-02-23 — Boston
- Public talk page: https://boston.aitinkerers.org/talks/rsvp_CMfZIqIUI5k

### [Canopy Agent Orchestrator: Everything Everywhere All at Once](https://nashville.aitinkerers.org/talks/rsvp_IyJRYGhJITc)

Canopy runs multiple coding agents in parallel, each in an isolated sandbox, with automatic merging and conflict resolution. It uses Beads for in-repo task tracking and dependency management, so agents work on the right things in the right order. The premise: engineering should be about specifying what you want, defining acceptance criteria, and reviewing results — not waiting for code to materialize one task at a time. Canopy treats code production as infrastructure that should get out of your way. I'll demo the system live: defining tasks with dependencies, launching parallel agents, and watching them work simultaneously while a dashboard tracks progress. We'll cover the copy-on-write isolation that makes this possible and how conflicts get resolved automatically.

- Event context: AI Tinkerers Nashville – January 29th, 2026: Live Demos, Code, and Architecture — 2026-01-29 — Nashville
- Public talk page: https://nashville.aitinkerers.org/talks/rsvp_IyJRYGhJITc

## Related Technologies

- [Claude Code](https://aitinkerers.org/technologies/claude-code) ([Markdown](https://aitinkerers.org/technologies/claude-code.md)) — 204 public demos
- [Go](https://aitinkerers.org/technologies/go) ([Markdown](https://aitinkerers.org/technologies/go.md)) — 24 public demos
- [Agentic coding CLI tool](https://aitinkerers.org/technologies/agentic-coding-cli-tool) ([Markdown](https://aitinkerers.org/technologies/agentic-coding-cli-tool.md)) — 1 public demo
- [bd mol](https://aitinkerers.org/technologies/bd-mol) ([Markdown](https://aitinkerers.org/technologies/bd-mol.md)) — 1 public demo
- [Beads (Go) Dolt (like Git for mysql](https://aitinkerers.org/technologies/beads-go-dolt-like-git-for-mysql) ([Markdown](https://aitinkerers.org/technologies/beads-go-dolt-like-git-for-mysql.md)) — 1 public demo
- [Canopy](https://aitinkerers.org/technologies/canopy) ([Markdown](https://aitinkerers.org/technologies/canopy.md)) — 1 public demo
- [Claude Agents SDK](https://aitinkerers.org/technologies/claude-agents-sdk) ([Markdown](https://aitinkerers.org/technologies/claude-agents-sdk.md)) — 3 public demos
- [Claude Opus](https://aitinkerers.org/technologies/claude-opus) ([Markdown](https://aitinkerers.org/technologies/claude-opus.md)) — 17 public demos
- [Codex CLI](https://aitinkerers.org/technologies/codex-cli) ([Markdown](https://aitinkerers.org/technologies/codex-cli.md)) — 5 public demos
- [compound-agent (npm)](https://aitinkerers.org/technologies/compound-agent-npm) ([Markdown](https://aitinkerers.org/technologies/compound-agent-npm.md)) — 1 public demo
- [Data](https://aitinkerers.org/technologies/data) ([Markdown](https://aitinkerers.org/technologies/data.md)) — 8 public demos
- [Dolt](https://aitinkerers.org/technologies/dolt) ([Markdown](https://aitinkerers.org/technologies/dolt.md)) — 1 public demo
- [DoltHub](https://aitinkerers.org/technologies/dolthub) ([Markdown](https://aitinkerers.org/technologies/dolthub.md)) — 1 public demo
- [frontier models (Claude](https://aitinkerers.org/technologies/frontier-models-claude) ([Markdown](https://aitinkerers.org/technologies/frontier-models-claude.md)) — 1 public demo
- [Gemini](https://aitinkerers.org/technologies/gemini) ([Markdown](https://aitinkerers.org/technologies/gemini.md)) — 187 public demos
- [GET](https://aitinkerers.org/technologies/get) ([Markdown](https://aitinkerers.org/technologies/get.md)) — 3 public demos
- [GitHub](https://aitinkerers.org/technologies/github) ([Markdown](https://aitinkerers.org/technologies/github.md)) — 73 public demos
- [git worktrees](https://aitinkerers.org/technologies/git-worktrees) ([Markdown](https://aitinkerers.org/technologies/git-worktrees.md)) — 2 public demos