Careers · Engineering
Applied AI Engineer, Evals
Make our agents measurably better. Build the evals, datasets and feedback loops that tell us whether a change improved scoring, research and outreach.
Israel
Hybrid
Full-time
About Clearitty
Clearitty is a team of AI agents that work a company’s market around the clock. They find accounts in motion, map the buying committee, plan the way in and run the outreach, and a person approves every step that matters. What sets us apart is the data underneath: real-world sales intelligence from thousands of sales executives, combined with buying signals and each customer’s own CRM. We’re a small team, backed by R-Squared Ventures, State of Mind Ventures and fresh.fund.
The role
Our agents make judgment calls: which accounts are in market, who really decides, what to say. You’ll define what good looks like for each one, measure it, and drive the improvements.
What you’ll do
Build evaluation suites for each agent: account scoring, stakeholder mapping, persona analysis and drafts
Create and maintain golden datasets from our sales intelligence and customer outcomes
Run experiments across prompts, models and retrieval, and ship the winners
Close the loop: feed approvals, edits and deal results back into scoring and agent behavior
Track quality, latency and cost per run, and catch regressions before customers do
What you bring
3+ years in ML, data science or applied AI engineering
Experience designing evals for LLM systems, including model-graded and human review
Strong Python and SQL, and solid statistics
Clear writing: you can explain why a number moved
Bonus points
Experience with ranking, scoring or recommendation systems
Familiarity with B2B sales data
How to apply
Click Apply to email us. Your email opens with the role already in the subject line. Add a few words about yourself and a link to your LinkedIn or your work.
Apply
