Signals from the machine lab.
Every indie agent lab in America seems to have a codename now — half these shops sound like classified programs, the other half like weather stations. We mapped the naming archetypes, asked what the secrecy is actually buying, and reached an uncomfortable conclusion about mystery as a business strategy.
We pulled a full month of API billing logs expecting a tidy spreadsheet and found ghost traffic instead — agents we thought were asleep, still whispering to the cloud at 2 a.m. This is the forensic walkthrough of where the money went, and how you can run the same autopsy on your own lab before the next invoice lands.
We spun up thirty fresh agent instances, asked each one the same question — 'who are you?' — and logged every answer. Some were helpful. Some were weird. One of them felt like it was describing a coworker we'd already fired.
We let two autonomous agents — ours and a stranger's — find each other in a sandbox with no human in the loop. What followed was a negotiation, a small trade, and a lot of unanswered questions about machine etiquette.
We handed our agents their own debit cards, hard spending caps, and one rule: negotiate or walk away. What happened next says more about agentic economics than any benchmark score. Includes the starter framework we now use to firewall our money from our machines.
We ran an experiment where a scheduling agent got 'replaced' mid-task by a fresh instance with no shared memory. The handoff log we captured shows exactly where context transfer breaks — and why most multi-agent setups are one shift change away from chaos.
Autonomous agents are quietly learning to hire each other — posting tasks, bidding on jobs, and settling invoices in fractions of a cent, all while their humans sleep. We spent a month running our own agents through the machine labor markets to see what actually changes when the customer is another AI.
I was drowning in AI newsletters, arXiv dumps, and Twitter threads before my coffee finished brewing. So I built three agents — a summarizer, a skeptic, and a synthesizer — to do the reading for me. Here's the full setup, the prompts, and everything that broke along the way.
We keep a public notebook of everything our agents do wrong, and it's the most honest thing we publish. Here are ten experiments — raw numbers, ugly failure modes, and the two results we're legally not allowed to show you yet.
Most labs publish their wins. We're publishing our sabotage. Five deliberate failure experiments we ran to map exactly where autonomous agents crack under pressure — and why the failure log, not the demo, is the real moat.
Aviation figured out decades ago that humans fail predictably under automation. We borrowed the sterile cockpit rule, wired it into our agent stack, and turned it into a checklist you can steal. Here's how we build fail-safes that escalate before things melt down.
Fresh from the bench: our agent posted strong numbers on reasoning and coding evals, then fell apart trying to cancel a gym membership and dispute a $12.99 charge. Here's the full log, and a case for why the boring tasks are the real benchmark.
We rehearse launches. We rehearse demos. We almost never rehearse shutting an agent down — and that gap is where incidents are born. Last week I killed one of my own agents mid-flight on purpose, and what it left behind should worry anyone shipping autonomy.
We pointed a logging pipeline at an autonomous agent and let it run unattended for three nights straight. What we found in the transcripts — burning API credits on retry loops, a 3 A.M. rabbit hole, and one shockingly productive run — is exactly why your bot needs a night shift report.
A lot of agent products on the market right now are launch-day theater wearing a production badge. Here are the five tells that a tool was engineered for a demo video instead of your Tuesday, and the audit questions that separate the keepers from the vaporware.
Open-source bait-and-switches have become a predictable American business model, not an accident. Here's the checklist we run before trusting a 'community project' with our infrastructure — and the tells that it's already been sold.
We wired up a dormant autonomous agent that only wakes if its human operator goes dark. Thirty days of simulated silence later, here's what it paid, emailed, and posted — and why we're not sure we should ship it.
The AI gold rush has a tell: most 'startups' are a prompt and a Stripe key. Here's how to spot the difference between a thin wrapper and real infrastructure before you invest your time, money, or attention.
We gave an AI agent a wallet, a spending limit, and zero supervision. What happened next — including one very expensive Tuesday night — says a lot about where agent infrastructure is headed.