Agent harness for multiple businesses
The client ran two businesses, DeFi growth and B2B services, on top of an open-source AI agent. We rebuilt that deployment into a production operations layer: high-volume automation with human-gated execution end to end, run by a single operator.
- ≈3.7Mtokens/day recovered from idle polling
- 892lifecycle sends/day re-lit within 48 hours
- <3daily notifications, down from 18–30
- 6,619upstream commits rebased behind a test gate
Challenge
One operator, two businesses, and an automation layer grown by accretion on a fast-moving open-source agent framework. The agent burned millions of tokens a day polling channels that produced nothing; dashboards showed 46/46 green while zero emails actually sent; six lifecycle email journeys sat dark for a week; the fork had drifted 6,600 commits behind upstream; and at 18 to 30 notifications a day, the operator had stopped reading any of them.
Insight
Every real defect shared one shape: components reported on themselves, not on their output. A green scheduler says the job ran; it says nothing about whether the email sent, the post published, or the money moved. Once health was redefined as verified business outcomes, and every irreversible action routed through a one-tap human gate, the machine could be trusted with volume while the operator kept judgment.
Approach
- Audited every scheduled job against its actual output; replaced always-on LLM pollers with deterministic gates that only wake the model when there is real work
- Replaced self-reported health with outcome contracts and canaries: checks that fail when the business result is missing, not when a process is
- Made upstream adoption routine with a 12-test safety net that fails on stock upstream, turning thousand-commit rebases into a scripted, reversible operation the client runs themselves
- Consolidated 18 to 30 daily pings into one actionable feed with a closed catalog of one-tap approvals; nothing publishes, sends or spends without the operator's tap
- Instrumented revenue truth end to end: person-level email-to-deposit attribution, measured sequence reply rates, and a daily call sheet sized to real capacity
Services
- AI Operations Audit
- Agent Harness Architecture
- Workflow Automation
- Approval & Safety Design
- Growth Instrumentation
Outcomes
- ≈3.7MAgent token burn eliminated from idle polling≈3.7M tokens/day recoveredVerified
- 892Dormant lifecycle emails re-lit892 sends/day within 48 hoursVerified
- <3Operator notification load18–30/day → under 3, all actionableVerified
- 6,619Upstream drift adopted safely6,619 commits rebased behind a test gate; updates now routineVerified
- 0False-green monitoring classEliminated; health now means output happenedVerified
- 4×Qualified lead pool4× expansion in one gated run (130 → 526 enriched leads)Verified
The client is a founder-led venture portfolio. All six outcomes carry a verified flag.
What changed
The harness now runs both of the client's businesses daily: content pipelines with validator gates, lifecycle email ladders under ramp caps, lead scoring on live hiring signals, and a morning call sheet that puts their operator on the phone with the twenty right people. Every send, post and spend still terminates in a human tap; the machine builds the lists, the operator does the talking.
Health is measured at the output: if the sends don't happen, the client knows the same day, from one feed rather than thirty pings. The safety net and update path were handed over as scripted, documented operations the client runs without us.