Software Engineer, Applied AI- Evals
MercorSan FranciscoOn-Site
Needs a free Superjobs account. You'll finish on Mercor's Ashby page.
Skill score
Summary
The Software Engineer on the Evals team builds scorecards, failure reports, dataset views, and self-serve onboarding that help customers run evaluations and act on results, and runs APEX benchmark launches. This suits a full-stack engineer with product sense who enjoys data-heavy products, customer work, and shipping iteratively.
- Function
- Development
- Estimated salary
- Est. $120k–180k/yr
- Estimated for a mid-level role, ~3 yrs experience (typical for the level), United States. Not published by the employer.
- Work mode and location
- On-Site, San Francisco
- Employment type
- Full-time
Job description
Overview
Mercor’s evals platform helps frontier labs and enterprises understand how their models perform on real work. Labs use it to decide what to train next, and enterprises use it to decide which models and agents to trust. It runs evals at scale, grades every result and explains why a model fails. The team also runs APEX, Mercor's family of benchmarks built with domain experts to measure how well frontier models do professional work in fields like law, finance and consulting. Its public leaderboard ranks the latest models, and its results appear in model cards across the industry.
As a Software Engineer on the Evals product team, you will build the scorecards, failure reports and dataset views that help customers turn eval runs into decisions. You will solve hard problems in scale, grading accuracy and reproducibility, while keeping the product simple to use.
This role grows with the platform, from evals into post-training and new product areas. You will help turn the failures we find into data that improves models.
Roles and responsibilities
- Build the product customers use to evaluate models: Ship the APIs, data models and UI that help customers run evals, compare models and share scores.
- Own APEX launches: Run every new frontier model on APEX and publish correct, defensible results to the public leaderboard.
- Explain why models fail: Build failure analysis that helps customers see where their models fail, with each pattern linked to the exact steps that went wrong.
- Make benchmarks easy to find and manage: Build dataset views, registries and versioning that help customers find the right benchmark or bring their own.
- Take customers from first login to self-serve: Build onboarding, roles, key management and usage views that help new teams run evals on their own.
- Work directly with customers: Join pilots with labs and enterprises and turn what they need into the roadmap.
Skills required
Top skills
- Python
- Go
- TypeScript
- React
All skills (6)
- Full-stack development
- Data models and APIs
- LLM evals
- AI benchmarks
- AI developer tools
- B2B developer products
Qualifications
- Bachelor's Degree in Computer Science or a related field
- Strong full-stack skills in a modern backend language (Python, Go or TypeScript) and a frontend framework such as React
- A track record of owning product features end to end, from idea to production
- Strong product sense that starts from what the user needs
- Experience building data-heavy products: data models, APIs and views over large volumes of results
- Excellent written and verbal communication with engineers, researchers and customers
- Comfort with ambiguity and a habit of shipping iteratively
Nice to have
- Experience with LLM evals, benchmarks or AI developer tools is a plus
- Startup experience or experience building B2B developer products is a plus
Other details
- Generous equity grant vested over 4 years
- Health, dental, and vision insurance
- 401(k) with company match
- Unlimited paid vacation and sick days
- Parental leave for birthing and non-birthing parents
- Free mental health services
- $200 monthly wellness reimbursement
- $1.5K monthly meal stipend
- $200 monthly laundry reimbursement
- Free Equinox membership
- Up to $15K relocation bonus
Company info
About Mercor
Mercor’s mission is to organize human intelligence for the AI economy. We build the infrastructure that enables frontier AI labs and enterprises to train, evaluate, and improve AI on real-world professional work. More than five million experts have joined our network, and we pay out over $5 million daily for