Careers
Work on AI that has to be right
We build and assure autonomous systems for finance, security, healthcare, telecom and government. If you want the proof to matter as much as the model, this is the work.
How we work
Small teams, hard problems
Each engagement is staffed by the few people who will build and assure it, end to end, with no hand-offs between the ones who design and the ones who verify.
Research that ships
Methods from the literature are tested against real systems, and what we learn goes back out as published articles, limits included.
Remote, written first
Decisions, specifications and findings are written down, so the work reads the same across time zones and holds up to an audit.
We're hiring
Current openings
Research Engineer, Interpretability
Build and validate probes, sparse autoencoders and circuit analyses that find failure modes before deployment, in models already carrying production traffic.
RemoteResearch
View role and how to applyAssurance Engineer, Formal Verification
Turn a client's operating constraints into machine-checkable properties, and prove them against agent behaviour before release, not after an incident.
RemoteAssurance
View role and how to applyApplied AI Engineer, Agentic Systems
Ship autonomous systems for regulated clients, with monitoring, escalation and a remediation lifecycle designed in from the start, not bolted on after the first incident.
RemoteEngineering
View role and how to applyRed-Team Engineer, Adversarial Robustness
Design attack suites for injection, escalation and goal hijacking against systems we are about to ship, and track regressions release to release so a closed hole stays closed.
RemoteAssurance
View role and how to applyAI Engineer, Autonomous Content Systems
Build and operate the autonomous pipeline that researches, writes, illustrates, fact-checks and publishes our own technical content, end to end, with no human in the loop on a normal day.
RemoteContent
View role and how to applyForward Deployed Engineer, Inference Infrastructure
Stand up the platform inside a client's own Kubernetes cluster or air-gapped network, and own the inference layer it runs on once it's there.
Remote, with client-site travelField Engineering
View role and how to apply
Don't see your role?
Write to us with what you would bring. We read every open application.
