About the role
You'll work on the AI surfaces of our product — the features that make our app feel smart instead of search-y. This is applied work, not research; success is measured in shipped features, latency, and customer wins.
Expect to move across prompting, retrieval, evaluation, and the messy glue between model and product.
What you'll do
- Build and ship AI features end-to-end — prompt, pipeline, eval, UI
- Stand up retrieval systems that actually retrieve the right thing
- Set up evals so we know when a change is an improvement vs. a regression
- Tune cost, latency, and quality across providers
- Work with product on which problems are worth solving with AI vs. not
- Stay current on what the field is doing — and what's hype
What we're looking for
- 2+ years shipping AI/ML features in a production product
- Experience with at least one of the major model providers (OpenAI, Anthropic, Google, open-source)
- Comfort with retrieval (vector DBs, hybrid search) and structured output
- A pragmatic streak — you know when to use a model and when to use a regex
Nice to have
- You've built evals for an AI product
- Experience with fine-tuning or distillation in production
- A demo we can play with
What we offer
- Senior-level salary plus equity
- A model-API budget that doesn't make you flinch
- Hardware on us
- Fully remote, with optional team off-sites
- A team that ships real AI features, not demos