I Automated My First Content Episode (Sort Of)
A rough first attempt at building a docs-to-audio content pipeline using Google Drive, NotebookLM, Gemini CLI, and Claude Code — honest notes on what it actually is.
5 min read
The flagship feed
AI in Living Color.
This is where I make sense of AI out loud. No hype, no doom, just the work and what it actually means. I track three things that are changing fast: the agents, the coding agents, and the tools for making content. The newsletter starts it, the show carries it, and short and long form take it everywhere else.
From the pulse · 2026-08-06
Live signal into Living in Color.
Last collected 2026-08-06 (14 days ago). Full list on /now.
- August AI Launch Calendar
Details Grok 4.6, GPT-6, Composer 3, Gemini 3.5 Pro, GPT Astra, Kimi k3.1, DeepSeek V4 GA for August.
- DeepSeek local open-weight security audit on Cake Wallet
Builder tests DeepSeek locally on-prem for PoC validation in Android emulator security audit; praises open-weight models for no-questions-asked tool use.
- Gemini 3.6 Flash - Model Card
DeepMind model card detailing OSWorld-Verified benchmarks, computer use capabilities, and integration with Google Antigravity.
- I ran Codex, Claude Code, Cursor, Hermes Agent, OpenCode, ZCode, and Grok Build through the same real development workflow
Builder details which agent wins for planning, coding, debugging, autonomous execution, and daily use after testing them side-by-side.
What I track
Three beats, graded on the Board.
Everything here runs off the same engine. The Bingo Board grades the models and harnesses, and these three beats are the lenses I read it through. 26 of 30 models graded so far.
Agents
The general-purpose agents the big labs are shipping. What they can actually run end to end, and where they fall over.
See the grades →Coding agents
The models and harnesses doing real engineering work, graded across five dimensions and re-synced three times a week.
See the grades →Content creation
The tools for making things. Writing, audio, video, the whole stack I use to ship in public.
See the grades →How it travels
One idea, every format.
It starts as writing, becomes the newsletter, then the show, then short and long form. Same thread, more rooms.
Newsletter
In developmentSignup is live. Weekly issues are not current. Latest archive is W22.
Podcast
In developmentThe show. Longer conversations and the beginner deep-dive series.
Vlogs
SoonBuilding out loud, on camera.
Minis
SoonShort, sharp clips for the feed.
Long form
SoonThe deep dives, the teardowns, the full builds.
Latest
Fresh off the feed.
Newsletter / 2026-W22
Latest issueAI This Week: 2026-W22
The week in AI: labs, models, and tools for coders and creators.
Voice AI Engineering · 12
How to Become a Voice AI Engineer: A Builder's Roadmap
Voice AI engineering spans ML, real-time systems, security, and product design—here's the exact phased learning path, skill map, and public build strategy being used to make the transition from security engineering to frontier voice AI.
10 min read
Voice AI Engineering · 11
Building AI for Black Communities and SMBs
Black communities shouldn't only be consumers of AI—they should be builders, owners, and beneficiaries: the case for starting with real community friction, not the latest model, and why data stewardship is a non-negotiable design constraint.
11 min read