CS-002 / B2B SaaS · customer support / AI audit
Found the agent quietly issuing $112K of unauthorized refunds.
An autonomous agent was resolving 38% of inbound tickets. Finance flagged a refund-rate anomaly; the support team couldn't explain it
// case studies · 31 engagements since 2023
Three engagements, told in full.
We have shipped 31 engagements. The referenceable work below is the work clients let us tell end-to-end. The rest lives behind NDAs.
// referenceable
Referenceable engagements.
Three engagements whose clients let us tell the story end-to-end.
CS-001 / Consumer lending / Eval-first build
An underwriting copilot that loan officers actually trust.
Twelve underwriters reviewing 800 applications a day. Average decision time was 14 minutes; senior staff were spending half their week on the easiest 60% of files
CS-003 / Logistics · enterprise / Embedded team
Stood up an in-house ML team and walked away when they were ready.
The internal team had built two models that worked in notebooks but had never made it to production. They needed a reference for what 'ready' looks like — eval surface, deployment patterns, on-call.
// confidential index
More work exists behind NDAs.
Healthcare diagnostics
Eval-first build
12 wks · Q4 '24
Anonymized clinical-decision-support agent
Procurement
AI audit
3 wks · Q1 '25
Pre-acquisition diligence on a target's AI claims
Defense contractor
Embed
Ongoing · '25
Tasking-and-orchestration system; under NDA
Financial markets
Eval-first build
16 wks · Q3 '24
Research-summarization for buy-side analysts
Legal tech
AI audit
4 wks · Q2 '25
Contract-review accuracy and privilege-leak audit
Energy sector
Embed
5 mo · '24
Anomaly detection on industrial telemetry
Public sector
AI audit
6 wks · Q1 '25
Procurement diligence; published version available on request
Insurance
Eval-first build
10 wks · Q4 '24
Claims triage with adjuster-in-the-loop