The company organizes human intelligence for the AI economy. It builds the infrastructure that frontier AI labs and enterprises use to train, evaluate, and improve AI on real professional work.
The company built its training and evaluation infrastructure with leading AI labs and is now offering those capabilities to enterprises in many industries. Its APEX benchmarks measure how AI performs on economically valuable work, and its enterprise evaluation and post-training tools turn those results into measurable gains in real workflows.
The work combines frontier AI, human expertise, and real-world use, and the problems are ones few companies have faced. The company looks for people who bring varied viewpoints, own their work, judge well, and hold a very high bar.
Staff work in person 5 days a week in San Francisco, New York, and London.
This is a Backend Software Engineer, Infrastructure role. You will design and build the core services and platforms that other engineering teams depend on, working with product, research, and operations engineers to solve hard scaling problems with well-designed, well-tested code.
Software engineering is the primary focus. Most of your time goes to designing systems and writing and reviewing code for services, platforms, and tooling that must run reliably in production at scale.
Hiring spans Platform, Developer Platform, and Storage and Databases. Teams are matched after the first screen, so apply even if your background points to one area more than another.
Responsibilities
Design and build multi-tenant platform services that give many teams isolated capacity, routing, quotas, and observability when they need it, and own them from architecture review through production.
Scale the gateway in front of every model provider the company uses, so internal teams and products get reliable, fair, cost-aware inference access without opening a ticket.
Build internal platforms and APIs that product and research engineers use daily, held to the same design and testing standard as customer-facing code.
Design data models, storage layers, and workflows on Postgres, MongoDB, and Temporal that can handle the next 10x of growth, including schema evolution, sharding and partitioning, and migrations.
Build developer platform components such as sandboxes, job execution, and service frameworks, where coding agents are treated as primary users alongside people.
Write design docs, lead architecture reviews, and set the standards for how services handle concurrency, retries, idempotency, and backpressure.
Improve the speed and efficiency of hot paths and make architecture changes that lower cost over time.
Requirements
The company cares more about how you reason about systems than which tools you have used. Strong candidates usually show the following.
Solid software engineering fundamentals: you design before you build, write code you are proud of in Python or Go, test it thoroughly, and review others' code carefully.
A deep understanding of distributed systems, including consistency, concurrency, failure modes, backpressure, idempotency, and capacity planning.
Experience designing APIs and services that other engineers consume, and changing them over time without breaking those consumers.
Solid experience with relational and/or document databases, including data modeling, query performance, and safe migrations.
Working knowledge of cloud infrastructure (the company runs on AWS) and containers. Deep Terraform or Kubernetes skills are not required, but you must be curious and quick to learn them.
High ownership: you spot a gap, write the proposal, ship the fix, and stand behind it in production.
You are a backend engineer drawn to the platform layer because that is where the hardest problems sit.
You have built internal platforms, SDKs, or shared services that other engineers enjoyed using.
You have worked on multi-tenant systems, rate limiting and quota systems, API gateways, or LLM and inference serving.
You have run databases, message queues, or workflow engines at meaningful scale and know their tradeoffs first-hand.
You have built developer experience tooling such as CI pipelines, build systems, local development environments, or internal CLIs that make engineers faster.
You are interested in developer productivity work alongside your main focus, such as cutting CI time, improving test reliability, or building sandboxes and guardrails for coding agents.
Conditions
The team is small and senior, with direct access to the Head of Infrastructure and engineering leadership.
In-person work 5 days a week in San Francisco. The company sees infrastructure as a team sport.
The company uses AI coding agents heavily and expects you to do the same. The main skill is deciding what the agents should do rather than typing the code yourself.
There is a weekly on-call rotation with structured handoffs, and the company actively works to reduce on-call load.
The company is an equal opportunity employer and does not discriminate based on protected characteristics under federal, California, or local law.
Benefits
Performance bonus paid bi-annually.
Equity grant vesting over 4 years.
Relocation bonus of up to $15k.
$10K housing bonus if you live within 0.5 miles of the office.
$1.5K monthly meal stipend.
Free Equinox membership.
$200 monthly laundry reimbursement.
$200 monthly personal wellness reimbursement.
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.
Where you’d work
From the office, 4 or more days a week in the office