/roles — ROLE_183
Software Engineer (SWE)
AI data startup building evaluation and training infrastructure used by top AI labs
The role
- COMP
- $260K - $290K
- EQUITY
- Competitive Equity
- LOCATION
- San Francisco
- WORKPLACE
- On-site
- EXPERIENCE
- 2 - 5 years
- VISA
- None, Visa transfers
- STACK
- JavaScript, TypeScript, React, NodeJS, Next.js, Python, Go, GCP, AWS, Kubernetes, Kafka, Redis, ElasticSearch
- INDUSTRY
- AI, Data
The company
Frontier-data company capturing how experts reason and turning real professional work into training data for foundation models.
- STAGE
- growth-stage
- FUNDING
- $30M+ raised
- TEAM
- ~80 people
- FOUNDED
- 2024
JD — the work
About the role
Engineers here own projects end to end across the stack, pushing forward how frontier AI systems are evaluated. The scope is wide: turning a research paper into a fresh RL environment, running distributed experiments on a Kubernetes cluster, making internal Next.js dashboards more reliable, or building Kafka streams that track annotator activity. It is a broad role rather than a narrow one, and it is on-site in San Francisco.
What you'll do
- Build APIs, RL environments, and scalable platforms that keep people in the loop
- Ship features fast alongside design, product, and research partners
- Keep the codebase readable and well documented
- Review teammates' code and weigh in on technical designs
- Tackle scaling and reliability problems in the core platform that labs rely on for data and evals
What they're looking for
- 2 to 5 years of software engineering experience
- Comfort across a stack that includes TypeScript, React, Next.js, Node, Python, and Go
- Experience with infrastructure such as Kubernetes, Kafka, Redis, Elasticsearch, and AWS or GCP
- Appetite for a wide-ranging role spanning research prototypes and production systems