Amplitude Wave Cuts Product Improvements to 1–2 Days With OpenAI Models
TL;DR: OpenAI published a customer story describing how Amplitude’s proactive product agent Wave uses OpenAI models to help teams ship product improvements in 1–2 days, down from roughly a month. The case matters because it shows a production agent routing work across reasoning, low-cost, and image models, then handing implementation plans to coding agents such as Codex.
What OpenAI published: a customer story about Amplitude’s Wave agent
OpenAI’s post is a customer story about Amplitude’s Wave product agent, not a new model launch. Amplitude describes Wave as roughly 80% powered by OpenAI models, including GPT-6.1 Sol, GPT-6 Luna, and GPT-Image-2.5 Flare.
Wave turns product insights into implementation plans that teams can send to coding agents, including Codex, and then measures the impact of the shipped change. Amplitude estimates that Wave can help teams ship product improvements in 1–2 days, down from roughly a month.
On Amplitude’s own homepage, Wave helped personalize suggestion cards, driving a 160% lift in card-driven sends.
“AI can now find useful work we never thought to assign it. OpenAI’s models can investigate a problem, write and run code, and use the results to decide what to try next.”
—Spenser Skates, CEO, Amplitude
How Wave works: from product data to shipped improvements
Wave is Amplitude’s proactive product agent. Teams give it a product area and the goals and metrics they want to improve. Wave then brings together Amplitude data, session replays, customer feedback, and code to identify and prioritize opportunities, develop solutions, and coordinate implementation with people and coding agents. It tracks outcomes and uses those learnings to inform the next round of improvements.
When an opportunity is ready to build, teams can send it directly from Wave to coding agents, including Codex, with a technical plan and acceptance criteria to guide implementation. The workflow is designed to connect product analytics and user behavior data to actual shipped changes rather than stopping at recommendations.
Model routing: different OpenAI models for different tasks
Amplitude assigns different OpenAI models to different parts of Wave based on the demands of each task. GPT-6.1 Sol supports the opportunity manager and deep investigations into code and data. GPT-6 Luna handles frequent, lower-cost tasks such as scanning code and running small evaluation checks. GPT-Image-2.5 Flare generates product mockups using screenshots from session replays as visual references.
The routing reflects a quality, latency, and cost tradeoff. A deep investigation needs strong reasoning and well-supported conclusions, while a small check repeated hundreds of times needs to be fast and inexpensive. Amplitude evaluates quality, latency, and cost for each task, then assigns models accordingly.
“OpenAI gave us both strong reasoning models and smaller, faster models, so we could balance quality, latency, and inference cost across the product”
—Spenser Skates, CEO, Amplitude
Evaluation results: GPT-6.1 Sol leads investigation and trades cost in data science
For Wave’s Opportunity Investigator, Amplitude evaluated whether models corroborated findings, avoided false positives, incorporated evaluator feedback, recovered from tool failures, and kept claims aligned with evidence. GPT-6.1 Sol scored 4.06 out of 5 overall, compared with 3.62, 3.52, and 2.94 for three alternative models. It led on all five investigation criteria and had the fastest median investigation time, at 54 minutes.
In a separate evaluation of Wave’s Data Science Agent, GPT-6.1 Sol scored 67.5 out of 100, compared with 69.3 for another leading model, at a mean model cost of $1.28 versus $2.63 per dispatch. GPT-6.1 Sol also scored higher on tool-contract compliance, estimation and uncertainty, and evidence grounding.
For generating wireframes and mockups, OpenAI’s image generation model scored 80.3 out of 100 for solution wireframes, compared with 63.9 for another image generation model. Amplitude’s evaluation covered communication quality, product fidelity, instruction adherence, text and UI legibility, and visual coherence across 15 opportunities.
Homepage experiment: 160% lift from personalized suggestion cards
Amplitude tested Wave on its own homepage, where generic suggestion cards were meant to help users start a conversation with an AI agent. Wave discovered that people often bypassed the cards: in a sample of 100 session replays, 80 users typed their own questions and only 12 selected a suggestion.
Wave proposed personalizing the cards using recent charts, dashboards, or the page a user had come from. It illustrated the change and produced a technical plan, including preserving free typing and showing generic suggestions when personalization context was unavailable. After a brief review, the team handed the work to a coding agent and A/B tested the change.
In the experiment readout Amplitude shared, messages sent through homepage suggestion cards increased 160% relative to control after one day.
“It is a great example of how non-groundbreaking things that just get ignored can increase the adoption of features in unexpected ways.”
—Eric Carlson, Senior Principal AI Engineer, Amplitude
Human control and launch speed
Teams keep control over what ships. Wave analyzes session and analytics data, then develops recommendations with a proposed fix, technical plan, and acceptance criteria. From there, teams can choose the level of automation. Most teams send bug fixes straight to coding agents, while new features and bigger bets get human review.
For changes that reach launch, the path from discovery to execution can be short. Among customer-launched experiments, flags, and guides, one in four went live the same day Wave surfaced the opportunity, half in under four days, and two in three within a week. The source notes that this curve is interpolated from supplied milestones and is illustrative, not individual observations.
What’s next
Amplitude and OpenAI frame the broader goal as making self-improving products a standard way companies build software. As OpenAI’s models get better at reasoning and coding, teams will be able to delegate more investigation and implementation work to agents, with the aim of getting better products to users faster.
“Our ambition with OpenAI is to make self-improving products the way companies build software.”
—Spenser Skates, CEO, Amplitude
Amplitude is already applying the workflow to its own product while the team works on other priorities. Skates adds: “I think teams will be surprised by how much they can accomplish when their software is working alongside them to make the product better.”