Mission
The world increasingly relies on AI agents, and those agents require enormous resources to run. Our mission is to make AI agents more efficient and, in doing so, enable teams to take on more ambitious problems.
Our first product is an automated context optimization service. It transparently optimizes agent trajectories to enable longer-running workflows, reduce token usage, and potentially improve task accuracy, all with zero code changes.
Join us
We describe our work as a cross between quant finance and AI engineering: hard optimization and infrastructure problems, measurable outcomes, and lots of room for new ideas.
If that sounds appealing, drop us a line at roles [at] induction.ai.
- + Induction
- Baseline
Each pair is one model and reasoning configuration: baseline and the same workload with Induction; bars show 95% confidence intervals. Induction shifts each configuration towards the lower cost quadrant.