• This is how I actually use AI agents to ship real features to production
  Watch the final outro segment to see a practical walkthrough of setting up and operating autonomous coding agents in production environments.


• Claude: Building Effective Agents
  An essential engineering guide by Anthropic explaining core patterns, design considerations, and architectures for building reliable agentic workflows.


• Stripe Minions: Stripe's One-Shot End-to-End Coding Agents
  A comprehensive multi-part whitepaper examining Stripe's methodology for implementing and utilizing coding agents in their ecosystem.


• CHUB: Accessing Up-to-Date API Documentation (by Andrew Ng)
  Discusses using agents with access to real-time, live API documentation instead of outdated training data, ensuring much higher quality output when combined with system rules.

• Agent Sandboxes
  A detailed exploration of secure sandboxing systems that provide isolated runtimes specifically tailored for executing agent code.


• Cloudflare Sandbox
  Learn how serverless environments and Cloudflare's platform can be adapted to serve as lightweight, secure execution layers for AI.


• Container Use
  Co-founded by Docker's creator, this explains using Docker containers as standardized agent tools (MCP), orchestrating them via Dagger, and diagramming the intersection of humans, LLMs, agents, and tools.

• How to Fix Vibe Coding (Syntax & Quality)
  Practical strategies for moving away from unstructured "vibe coding" by enforcing deterministic tooling, compilers, and validation.


• How to Do It Like a Professional
  A deep dive into professional software engineering practices and organizing developer workflows to ensure high-quality AI code generation.


• Claude Code Review
  Demonstrates how to integrate automated LLM reviews to check syntax, structural soundness, and logic errors.


• Be Skeptical Without Being Cynical
  A prompt looking at why seeking deep insights is significantly more critical than just amassing raw information and knowledge.


• Real-World Agent Vulnerabilities & Leaks
  A case study on malicious GitHub issues exploiting agent tool permissions to leak proprietary code, stressing the need for human validation.

Development & Analysis

• fallow.tools: Deterministic analyzer for TS/JS checking for dead code, duplication, circular dependencies, and complexity.
• Project Wallace: CSS analysis tool to monitor design system bloating and complexity.
• ESLint & Stylelint: Deterministic checkers for JS and CSS that prevent the AI from bypassing structural and formatting constraints.
• Vite Plus (vp check): Added build check integration for static and dynamic correctness.

Runtimes & Component Libraries

• Storybook: Component building, viewing, and visual unit testing.
• Sentry: Bug tracking and crash reporting in production.
• Agent Browser & Chrome DevTools MCP: Tools for headless browsing, capturing screenshots, and visual page validation.

Workflows & Documentation

• dex.rip: Structured, clear workflow tracker for executing software tasks.
• beads: Git-hook integrated database ensuring clear execution order for dependent tasks.
• context7: Retrieves and indexes the latest API docs directly to provide fresh patterns and context to the LLM.
• TanStack Code Mode: Running TypeScript code dynamically inside isolated execution sandboxes.
• GSD Redux: One of the core orchestrators used to manage and coordinate the agentic execution process.