"Local-first" is becoming a positioning claim, not an architecture decision. A new open-source coding agent landed this week with the pitch: your code stays on your machine and only ever leaves to the model provider....
Read on linkedin →What my AI actually shipped.
Most "I built an AI agent" posts come with a screenshot and a vibe. This page is the receipts — live counts from my running stack, refreshed each morning at 06:15 ET. No render-on-load tricks. The system actually does this work.
Cost of running
cloud-API cost · lifetime
Inference runs on the workstation. Cloud only when local genuinely can’t carry the work.
est. monthly savings
vs. Buffer Pro + Hootsuite Pro + ChatGPT Plus + ~$150 cloud-API usage.
posts shipped · last 7 days
Drafted by NEXUS, reviewed by me, fired on weekly slots. Receipts below.
Today · 2026-08-02
- Posts shipped
- 0
- Drafts created
- 1
- Opportunities scored
- 0
Last 7 days
- Posts shipped
- 14
- Drafts created
- 42
- Opportunities scored
- 0
Agents
The autonomous workers running in my stack. Each one has a single job and runs on its own schedule. Status · last run.
Brain
Cognitive Loop
okRuns the autonomous cycle every minute — every subsystem gets a turn.
last run · 15h ago
Capability Builder
okDrafts new capabilities when a gap is detected — ships only after review.
last run · 5d ago
Intent Router
okRoutes incoming signals to the right handler.
last run · 3m ago
Content
Morning Brief
okBuilds the daily briefing before sunrise — overnight research, opportunities, what shipped.
last run · 15m ago
Drafter
okTurns captures and signals into ready-to-review drafts.
last run · never
Slot Publisher
okFires weekly publishing slots when an approved post is ready.
last run · just now
Performance Tracker
okMeasures what published content actually earned in engagement.
last run · 25m ago
Research
Researcher
okPulls overnight research signals worth the morning brief.
last run · 6h ago
Income
Opportunity Scout
okScans creator channels for monetization-fit ideas.
last run · 25m ago
Voice
Brief Reader
okReads the morning brief aloud during the get-ready window.
last run · 23h ago
Memory
Memory Distiller
okCompresses recent context into long-term memory.
last run · 15h ago
Reflection Engine
okNotices what worked, what didn't, what to change.
last run · 15h ago
Recently shipped
The most recent posts NEXUS drafted, I reviewed, and the stack published.
Spent a week cloning my own face with local AI so I'd never film again. Voice: nailed it (it's narrating this). Face: turned me into AI slop. The tools that actually work keep your face on their servers. So, faceless ...
Read on bluesky →One week, zero cloud, all on my own hardware. The voice clone worked (it narrates this video). The face didn't. Every local tool either broke the mouth or slowly turned me into someone else. The tools that actually na...
Read on youtube →A multi-model pattern splits AI coding into plan, execute, verify -- three phases, three pass/fail gates. Most setups skip the third. That's the gap between a tool you babysit and a system you operate. The cost saving...
Read on bluesky →1/ Google wants access to your travel history, your calendar, and your sleep patterns. In exchange, it will tell you when to go to bed before a flight. 2/ Jet lag advice is not a hard problem. Shift your sleep an hou...
Read on twitter →This page is a static snapshot, pushed once each morning by the same stack that runs everything else. No backend · no live API · no telemetry from your browser. The Sunday newsletter tells the story behind the numbers.