Classifier
Score a lead based on your team’s criteria.
Sutro is a platform for building, optimizing, and running reliable AI Functions.
A reliable AI Function repeatedly and consistently makes the same decisions as a human expert would on a given task.
Book a demoAn AI Function is a repeated AI task. It performs the same operation, many times, over varying inputs. Below are examples of AI Functions our customers are building.
Score a lead based on your team’s criteria.
Give pass or fail assignments on agent traces.
Remove personally identifiable information from medical records.
Determine whether two businesses are the same.
Determine whether a candidate is a good fit for a role.
Send support tickets to the right team or person.
Building an AI Function isn’t about scaling general intelligence. It’s about teaching a model exactly how your organization wants a task performed.
Sutro fine-tunes your prompts for any given task. Upload an unlabeled dataset, provide feedback on a few hard cases, and Sutro optimizes a prompt that will reliably execute your task using off the shelf models.
Try it out below.
Is this hacker news post related to aviation?
Hacker News post
Sutro can now generalize the rules behind your feedback.
Run a function continuously as new inputs arrive, or process large datasets in batch.
Sutro is powering critical, production AI workloads with many happy customers today. Case studies are available upon request.
Agents, evals, matchers, and other production AI workflows.
Data enrichment, entity resolution, classification, and extraction.
Data filtering, labeling, tagging, and dataset preparation.
Request customer case studies →“Sutro saves our team countless hours, and it gives us the invaluable ability to measure, optimize, and prevent regressions against our domain expertise. It’s a must-have for serious AI developers.”
$500 / month
(includes $100/mo inference credits)Enterprise and self-hosted plans are tailored to your needs and scale.
See full pricing →Machine-time pricing means you only pay for the compute your workload actually consumes. Sutro is surprisingly affordable for large datasets.
Yes and no. Sutro helps evaluate quality and reliability, but unlike other evals products, it directly influences model/agent behavior from feedback.
Yes. Sutro supports images, PDFs, and web search capabilities today. Additional modalities and custom tool calling are available on a per-request basis.
Functions are typically model-agnostic, and we support a mix of open-source and proprietary models. It's common that small-open source models outperform larger proprietary models when building AI Functions. Sutro helps automatically find the best model for your needs.
Yes. Sutro can be managed as SaaS or entirely self-hosted. You can bring your own provider keys and cloud credentials in either case.
AI Functions accumulate edge cases by nature of the volume they run at and the diversity of possible inputs. Sutro systematically discovers these edge cases and learns generalized rules optimized to cover as many of them as possible. Manual prompt engineering is typically not scalable in these scenarios.
Sutro Functions are designed to be extremely data efficient and portable. Functions learn in-context, meaning no weight updates are needed and can be continually optimized. Fine-tuning and reinforement learning typically require more data and require retraining for updates, making them more rigid and less portable.