IDEAS, DECISIONS & LESSONS

Journal

Notes on AI, architecture and building useful products.

26 articles

10 Sep 2026 · 5 min read

agentic coding

10,000 Agents, 88 Hours, One Disputed Proof: What OpenAI's Navier-Stokes Claim Means for Engineering Teams

OpenAI says an internal model coordinated 10,000 agents for 88 hours to solve the Navier-Stokes Millennium Prize problem — then a mathematician accused it of building on unpublished work it had access to. A technical lead's take on the swarm-scale mechanics and the attribution problem every AI-assisted team will eventually hit.

27 Aug 2026 · 4 min read

agent architecture

How xAI Builds Grok — and How I'd Build a Grok of My Own

Grok isn't a model, it's an agent system: tools trained into the reasoning, real-time X and web search on xAI's infra, a four-agent cross-verifying architecture, long-running agents. I break down its features, architecture, and safety harness — then map each piece to a component I can build self-hosted on a subscription I already pay for.

24 Jul 2026 · 5 min read

OpenAI Codex Voice + Multi-Agent: What the July 2026 Announcement Means for Your Codebase

OpenAI just shipped Codex with voice control and parallel agent subprocesses. Here's an honest Tech Lead breakdown of what's actually new, how it compares to Claude Code and GitHub Copilot Workspace, and whether voice-directed coding is a gimmick or a genuine productivity multiplier.

18 May 2026 · 11 min read

The Multi-Agent Coding Grid: How I Use Claude as an Orchestrator with MiniMax and Kimi as Workers

A practical guide to running parallel AI coding agents using Claude as a manager, MiniMax and Kimi as workers, Linear as the task pool, and tmux as the control room. No frameworks — just shell scripts, lock files, and clear task specs.

20 Apr 2026 · 18 min read

The SSE Agent: Code Generation, Self-Testing, and Iteration (Part 7 of 12)

Part 7 of 12 — Build the Senior Software Engineer agent that writes code, runs tests, and iterates until they pass.

20 Apr 2026 · 15 min read

TL, DevOps, and PM Agents: Review, Deploy, and Coordinate (Part 8 of 12)

Part 8 of 12 — Build the Tech Lead (code review), DevOps (CI/CD generation), and PM (coordination) agents.

20 Apr 2026 · 20 min read

The Complete Workflow: Wiring All Agents End-to-End (Part 9 of 12)

Part 9 of 12 — Assemble all 8 agents into a single LangGraph workflow. Full E2E example from client brief to deployed config.

20 Apr 2026 · 15 min read

agent-learning

Agent Learning: Memory, RAG, and Skills That Grow Over Time (Part 10 of 12)

Part 10 of 12 — Make agents smarter over time with long-term memory, RAG, task reflection, and inter-agent knowledge sharing.

20 Apr 2026 · 14 min read

Dashboard: Real-Time Visibility Into Your AI Team (Part 11 of 12)

Part 11 of 12 — Build a real-time Streamlit dashboard: Kanban board, agent status, cost tracking, and human checkpoint approval.