hub / github.com/ai-that-works/ai-that-works
github.com/ai-that-works/ai-that-works @main sqlite
2,767 symbols
8,638 edges
667 files
781 documented · 28%
README
🦄 AI That Works
On Zoom, Tuesdays at 10 AM PST - an hour of live coding, Q&A, and production-ready AI engineering
🦄 Next Episode
Product Specs with AI
Tuesday, June 16, 2026 at 10 AM PST
We've talked a lot about design discussions for planning work with AI and getting leverage before writing the code, but this process has a common pitfall: it combines product decisions (how does it work, what is the user experience) with technical decisions (how do we build it, what patterns do we follow).
This complecting of concerns can cause important questions to be missed. On today's AI that works we'll dig into techniques to split out product vs. technical questions to enable less-technical folks to participate in product specification that is grounded in codebase research, and ensure technical depth is achieved without getting distracted by product questions.
What We're About
Weekly conversations with @hellovai & @dexhorthy about getting the most juice out of today's models
When: Every Tuesday at 10 AM PST on Zoom
Duration: 1 hour of live coding, Q&A, and production-ready insights
Goal: Take your AI app from demo → production
Let's code together.
Pre-Reading & Setup
Before joining, get familiar with our toolkit:
| ### **Core Tools** - **Zoom** - Live sessions - **Cursor** - AI-powered IDE - **Git** - Version control - **Claude Code** - Agentic Coding - **CodeLayer** - Agentic Coding Tool | ### **Languages** - **Python/TypeScript/Go** - Application logic - **BAML** - Prompting DSL - [Repository](https://github.com/boundaryml/baml) - [Getting Started Guide](https://gloochat.notion.site/benefits-of-baml) | ### **Package Managers** - **Python:** [UV](https://docs.astral.sh/uv/getting-started/installation) - **TypeScript:** PNPM - **Go:** Go modules |
Episodes & Workshops
From Demo to Production - One Episode at a Time
| 📅 Episode | 📝 Description |
|---|---|
| UPCOMING 2026-06-16 #62: Product Specs with AI code • register | We've talked a lot about design discussions for planning work with AI and getting leverage before writing the code, but this process has a common pitfall: it combines product decisions (how does it work, what is the user experience) with technical decisions (how do we build it, what patterns do we follow). This complecting of concerns can cause important questions to be missed. On today's AI that works we'll dig into techniques to split out product vs. technical questions to enable less-technical folks to participate in product specification that is grounded in codebase research, and ensure technical depth is achieved without getting distracted by product questions. |
| PAST 2026-06-09 #61: Hands-on with Fable 5 watch • code | We had agent observability on the schedule, but Anthropic shipped Fable 5 about twenty minutes before we went live, so we tossed the plan and got hands-on with the new model instead. Zero prep, all live. This is an unscripted look at exactly how we kick the tires on a fresh model release: take the hardest problem you're already deep in, hand it over, and watch whether it finds leverage you didn't. Vaibhav ran it against an in-progress design doc for the BAML VM's observability layer, Dex ran it against an old race-condition benchmark he keeps around, and we talk through what the model caught, what it missed, and why most model releases are more hype than the value they deliver day to day. |
| PAST 2026-06-02 #60: How to Build AI Agents that Work in Any Language watch • code | In this episode, we discuss the challenge of building multilingual AI applications that perform consistently whether your user is interacting in English, Spanish, French, whatever. Can you simply run an English prompt through a basic translator? Or will that break down in production? We'll be breaking down practical engineering strategies for designing flexible, cross-lingual prompt architectures that maintain semantic alignment without forcing you to build and manage separate pipelines for every language. |
| PAST 2026-05-26 #59: No Vibes Allowed: Performance Engineering code | This week on the podcast, we are doing another no vibes allowed episode focusing on performance engineering. We will dive into high-performance engineering on virtual machines to help you maximize efficiency out of your compute infrastructure. When you are deploying heavy LLM workloads or complex pipelines, an unoptimized VM can lead to crippling latency, memory bottlenecks, and ballooning costs that break your application in the real world. We break down practical techniques to configure your environments, manage resources, and eliminate overhead so your models run flawlessly under pressure. |
| PAST 2026-05-19 #58: How AI Agents Can Safely Ship Code to Production watch • code | This week, the top headline is vibe coders realizing that they can use feature flags to ship experimental (read: slop) features to production without impacting all customers. Shipping code is a lot harder when everything is changing all the time. Feature flags can be a good technique to test various things, but how do you set that up? Do you feature flag new models? New prompts? New harnesses? We'll dive into details here and see where feature flags improve your product delivery vs. just giving you an excuse to ship more slop. |
| PAST 2026-05-12 #57: "Code Mode" Deep Dive watch • code | On Monday, Pash from OpenAI shared that Codex has a secret "code mode" feature - an alternative to traditional tool calling. There's a lot of debate going on around the best way to give tools to models - skills vs. mcps, CLIs and bash vs custom tools, or letting the model write code for everything. In this episode we're going to cut through the hype and dive deep on the differences and tradeoffs between these methods. • What is "code mode" and how does it work • Tradeoffs between MCP vs. Bash+CLI vs. Code mode • Why it matters to agent or harness builders |
| PAST 2026-05-05 #56: OpenAI tells you not to build your own harness watch • code | Harness engineering is all the hype now, so on this week on the podcast we're looking back to an article written by OpenAI in February about harness engineering, "Harness engineering: leveraging Codex in an agent-first world". In this article, they claim that the era of "hand-written code" is officially over. We break down their experiment of shipping a million-line product with zero manual coding, shifting the human role from "coder" to "environment designer." |
| PAST 2026-04-28 #55: No Vibes Allowed - Building Design Docs with AI watch • code | In this month's no vibes allowed episode, Vaibhav will show how he uses AI to make design docs for complicated tasks by building out an actual design doc for a feature in BAML. As always for our no vibes allowed series, we will be solving real problems in real production systems. |
| PAST |
Extension points exported contracts — how you extend this code
ThreadStore (Interface)(no doc) [9 implementers]
2025-06-03-humans-as-tools-async/src/state.ts
ParserState (Interface)
* Parser state for incremental parsing
2025-11-05-event-driven-agents/demo/src/antml/AntmlParser.ts
LumaEvent (Interface)
(no doc)
tools/luma.ts
ZoomToken (Interface)
(no doc)
tools/zoom.ts
ValidationResult (Interface)
(no doc)
tools/validate-metadata.ts
KeyboardShortcutProps (Interface)
(no doc)
2026-04-14-agentic-coding-for-frontend-apps/02-storybook-riptide/src/components/keyboard-shortcut.tsx
User (Interface)
(no doc)
2026-04-14-agentic-coding-for-frontend-apps/03-wired-vs-pure/src/types.ts
Notification (Interface)
(no doc)
2025-10-28-ralph-wiggum-coding-agent-power-tools/webapp/src/components/notifications/NotificationList.tsx
Core symbols most depended-on inside this repo
getcalled by 514
2025-06-03-humans-as-tools-async/src/state.ts
error
called by 313
2025-09-23-evals-for-classification/src/shared/logger.py
log
called by 168
2026-01-13-applying-12-factor-principles-to-coding-agent-sdks/src/utils.ts
info
called by 112
2025-09-23-evals-for-classification/src/shared/logger.py
exists
called by 106
2026-01-13-applying-12-factor-principles-to-coding-agent-sdks/src/store/order-store.ts
split
called by 87
2026-04-11-unconf-sf/src/talk_segmenter/transcript_splitter.py
write
called by 87
2026-04-11-unconf-sf/src/talk_segmenter/segment_writer.py
open
called by 86
2025-11-05-event-driven-agents/demo/src/server.ts
Shape
Languages
TypeScript60%
Python40%
Modules by API surface
2025-07-01-ai-content-pipeline-2/backend/main.py49 symbols
2025-09-23-evals-for-classification/tests/unit/classification/vector_store_test.py38 symbols
2025-06-24-ai-content-pipeline/backend/main.py37 symbols
2025-10-21-agentic-rag-context-engineering/tui.py32 symbols
2025-09-23-evals-for-classification/tests/unit/classification/embeddings_test.py32 symbols
2025-09-23-evals-for-classification/tests/unit/classification/selection_test.py28 symbols
2025-10-28-ralph-wiggum-coding-agent-power-tools/webapp/src/lib/email-notifications.ts27 symbols
2025-08-19-interruptible-agents/runtime.py27 symbols
2025-12-02-multimodal-evals/src/receipt_evaluator.py25 symbols
2025-10-28-ralph-wiggum-coding-agent-power-tools/webapp/src/app/actions/todos.ts25 symbols
2025-07-01-ai-content-pipeline-2/backend/models.py25 symbols
2025-09-23-evals-for-classification/tests/unit/classification/narrowing_test.py24 symbols
Dependencies from manifests, versioned
@anthropic-ai/claude-agent-sdk0.2.38 · 1×
@biomejs/biome2.2.0 · 1×
@boundaryml/baml0.88.0 · 1×
@boundaryml/baml-nextjs-plugin0.1.0 · 1×
@chromatic-com/storybook5.1.2 · 1×
@dagrejs/dagre1.1.5 · 1×
@eslint/eslintrc3 · 1×
@hono/node-server1.13.7 · 1×
@hookform/resolvers5.1.1 · 1×
@linear/sdk1.22.0 · 1×
@mdx-js/mdx3.1.0 · 1×
@prisma/client6.18.0 · 1×
For agents
$ claude mcp add ai-that-works \
-- python -m otcore.mcp_server <graph>