AI toolsTrendingmedium impact
Amplitude's Wave agent ships product changes in 1-2 days using OpenAI
Amplitude built its proactive product agent, Wave, primarily on OpenAI models, and reports it now ships product improvements in an estimated 1-2 days. The company also reports a 160% lift in agent engagement versus control after one day, a result that shows how quickly an AI-driven build-measure loop can change a product.
Key Facts
01Amplitude's agent Wave is built primarily on OpenAI models and covers identifying, shipping, and measuring product changes.
02Amplitude estimates it can ship a product improvement in 1 to 2 days.
03The company reports a 160% lift in agent engagement versus control after one day.
04The case study does not detail the size or complexity of the changes behind the estimate.
Why does this matter for agencies?
▶Shorter build-measure cycles let agencies run more client experiments each month.
▶Measurement is what turns faster shipping into useful results, not just more output.
▶Clean analytics tracking becomes a prerequisite for any agent-driven testing.
What should agencies do?
Map the days between a test idea and a live change for one active client to set a baseline.
low effort
Pilot a short build-measure loop on one low-risk page with a single success metric.
medium effort
Audit event tracking and set human approval gates for AI-written changes before going live.
medium effort