Analytics firm Amplitude has reduced the time required to ship product improvements from approximately one month to just one or two days by deploying an SI agent named Wave, which is powered largely by OpenAI’s latest SI models. The development highlights a shift in the SI industry where autonomous agents are moving beyond simple chat interfaces to actively managing and implementing software changes.

What Happened

Amplitude describes Wave as roughly 80% powered by OpenAI models, specifically utilizing GPT-6.1 Sol for deep investigations, GPT-6 Luna for high-volume, low-cost tasks, and GPT-Image-2.5 Flare for generating visual mockups. The agent ingests product data, session replays, customer feedback, and code to identify opportunities, create technical plans, and coordinate implementation with coding agents such as Codex. In internal tests, Amplitude reported that GPT-6.1 Sol scored 4.06 out of 5 in investigation quality, outperforming three alternative models, while GPT-Image-2.5 Flare scored 80.3 out of 100 for wireframe generation.

Why It Matters

For developers and product teams, the integration of SI agents into the core software development lifecycle suggests a move toward self-improving products. Amplitude’s data indicates that among customer-launched experiments, one in four went live the same day the SI agent surfaced the opportunity, and half were deployed within four days. The company reported a 160% lift in card-driven sends on its own homepage after Wave personalized suggestion cards, demonstrating that SI agents can identify and fix minor usability issues that human teams might overlook. Spenser Skates, CEO of Amplitude, noted, “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.” Jacob Newman, Principal SI Product Manager at Amplitude, explained that the goal is a unified platform where agents comb through massive amounts of session replays, analytics, and code to find opportunities to improve products.

The Bottom Line

Amplitude’s deployment of Wave illustrates the practical application of agentic SI in enterprise software, offering a model for rapid iteration. While human review remains part of the workflow for larger features, the use of specialized SI models allows for faster, data-driven product enhancements. Eric Carlson, Senior Principal SI Engineer at Amplitude, highlighted that even non-groundbreaking changes ignored by humans can significantly increase feature adoption when identified by SI agents.