Your next chapter.
Build AI features people can use with confidence. From source-linked summaries to assisted support and product recommendations, your work will connect model capabilities to clear customer outcomes.
You’ll own the path from an experiment to a dependable experience: choosing the right task, evaluating performance, understanding failure modes, and giving people appropriate control. Curiosity matters, and so does knowing when a simpler solution is better.
What you’ll do
- Prototype retrieval, summarization, and assistance features
- Evaluate quality, latency, and cost against real tasks
- Design source attribution and human review controls
- Create evaluation sets based on realistic customer tasks.
- Design retrieval, grounding, and review workflows with the product team.
- Monitor quality and cost, investigate failures, and document limitations.
What you’ll bring
- Experience building and evaluating AI applications
- Strong software engineering fundamentals
- Good judgment about where automation belongs
- Experience with model evaluation and retrieval-based applications.
- Ability to investigate data quality, failure cases, and ambiguous results.
- Comfort measuring latency, cost, and user value together.
A strong start looks like this.
In your first few months, we’d focus on learning together and making a useful contribution.
- 01
Identify a focused task where AI offers a measurable advantage.
- 02
Ship an evaluated prototype with sources and human review where needed.
- 03
Establish a repeatable quality baseline for the next iteration.
Good work happens together.
You’ll collaborate with product researchers, full-stack engineers, design, and customer-facing specialists. We value clear writing, open feedback, time to focus, and practical decisions made close to the work.
Bring your perspective, curiosity, and care for customers. We’re interested in how you think, what you’ve learned, and the difference you can make with a team.