Careers at Solumn AI
Build the safety layer
of frontier AI.
Red teaming, vulnerability discovery, evaluations and the datasets that make advanced models safe to ship — the assurance layer the frontier labs depend on, built to a benchmark standard. We're hiring across it.
The work
If capability asks what a model can do, we ask what happens when it fails.
We work with leading AI labs to make advanced models deployment-ready, across the post-training and assurance stack. Every technology cycle begins with capability and matures through trust — the internet, cloud, mobile and social all ran that arc. AI is crossing that threshold now, and the assurance layer is becoming foundational.
Applied safety research
Open questions in alignment, misuse and agentic risk, turned into the taxonomies and evaluation methods the rest of our work is built on.
Red teaming & evaluation
Adversarial attacks on tool-calling and agent deployments, scored against a harm taxonomy.
Vulnerability & secure code
Find-and-fix security tasks from real CVEs, each backed by a verifier that runs a real exploit and regression test.
Agentic environments
Containerised tasks and deterministic graders that test how an agent behaves under pressure.
Open roles
Who we're hiring.
All our roles are remote. Expect a short practical task and a couple of focused conversations — we try hard to respect your time.
Head of Talent Acquisition (Global)
Build and lead our talent function from the ground up — the global network of freelance engineers and domain experts who power our client work, and full-cycle hiring for the core team. Hands-on leadership, reporting to the founders.
Team Lead / Quality Calibrator
Set the quality bar for everything we ship. You author tasks and reference solutions, lead QC, and run the calibration and delivery around it — reviewers, guidelines and metrics.
AI Engineer
Train and fine-tune models, and own the platform, benchmarks and scale behind our work — Dockerized environments, graders and rollout pipelines. Hands-on, with engineering-manager scope.
LLM Trainer — Code Security
Judge how AI coding agents behave on real software-engineering and security tasks — whether a result is genuinely correct and secure — and repair the tasks that are wrong. High-judgment review work, a strong on-ramp into frontier AI safety.
Don't see your exact role but know you can contribute here? Apply below and choose "Open application" — or write to hello@solumn.ai.
Life at Solumn
A small team, doing work that matters immediately.
You own real problems from day one, alongside the people whose work the frontier labs depend on. We're honest about the pace, and we invest in the people who keep it.
How we work
Remote, everywhere
Every role is remote. Work from wherever you do your best work — no relocation, no commute.
Flexible hours
Own your schedule around your own rhythm. We care about what you ship, not hours logged at a desk.
A six-day week
We're early and we move fast, so we work six days a week. Remote work and flexible hours are how we keep that sustainable — we're upfront about it so you can choose with eyes open.
What you get
Health cover for your family
Group health insurance that covers you, your spouse and your children — not just the employee.
Work that ships to the frontier
What you build is reviewed, graded and used by the labs training the largest models that exist.
Grow into a rare craft
Few people do this work. You'll learn it directly, with mentorship and room to lead as we scale.
Apply
Submit your application.
Tell us who you are and what you've built or led, and pick the role you're applying for. We read for evidence over credentials.
Still a student?
The Foundations Programme is our paid engagement for final-year students across the IITs, NITs, BITS and IIITs — build real environments on live client work, with a pre-placement offer on merit.