Careers · Talent pool
Make AI behavior understandable and usable. This role connects product design, interface engineering and the evaluation of what a model returns to a person.
This is a talent pool, not an active vacancy. Share your portfolio for consideration when a suitable opportunity becomes available.
Why this role
A model response is only one part of an AI product. People also need to understand what is happening, judge the result and decide what to do next.
Products and tasks
AI assistants, copilots, generative workflows and review tools within SaaS and enterprise products. The focus is the user-facing AI scenario, not simply using AI to write code.
Prototype and implement conversational flows, streamed responses, progress feedback and recovery from interrupted or failed requests.
Design ways to edit inputs, inspect outputs, correct mistakes and approve or stop an action before it has consequences.
Define representative scenarios and quality checks with the team. Use observed failures to improve the interaction and its boundaries.
React and TypeScript experience with asynchronous APIs, streamed data and interface state management.
An understanding of model uncertainty, response variability and the difference between generating a result and validating it.
The ability to translate a user task into an AI-assisted flow with clear controls, accessible feedback and useful error states.
Retrieval, tool-calling or evaluation workflows that you have connected to a user-facing product.
Usability research for AI experiences, source attribution, feedback collection or review interfaces for sensitive decisions.
An interactive AI workflow, with both the intended path and examples of uncertain, incomplete or incorrect output.
Show how users can interrupt, revise, review or recover. Explain what remains under human control.
Include the scenarios or evaluation criteria you used and what changed after testing. A generated screen alone does not show this work.
Work with product designers and the people responsible for models, data and APIs. Review interface behavior and output quality together.
Before a future engagement, define the use case, data constraints, decision boundaries and responsibility for evaluating results.
Share your name, email, portfolio and CV as a file or link. Explain your contribution to the work you show and the discipline you would like to work in.
This is a talent pool, not an active vacancy. Share your portfolio for consideration when a suitable opportunity becomes available.