Thinking Machines Lab jobs

Network Engineer role at AI/ML research lab focused on large-scale GPU cluster networking, RDMA/RoCE fabric debugging, NVLink management, and infrastructure instrumentation. Full-time on-site position in San Francisco requiring backend engineering expertise in Python or Rust.

Thinking Machines Lab is hiring a Site Reliability Engineer for Tinker, their fine-tuning API platform that lets researchers customize frontier AI models. You'll own end-to-end reliability across the distributed training infrastructure, including CI/CD pipelines, production observability, incident response, and multi-tenant resource isolation for GPU workloads. The role requires a CS degree or equivalent, proven experience with distributed systems and cloud infrastructure, strong software engineering skills for building reliability tooling, and a track record of production incident management—preferred qualifications include operating large-scale cloud services, experience with distributed training frameworks, and Kubernetes expertise at scale. Based in San Francisco, the position offers $350k–$475k annual compensation plus equity, generous benefits, unlimited PTO, and visa sponsorship. This is a hands-on infrastructure role supporting a rapidly scaling platform with novel use cases in AI model fine-tuning.

Thinking Machines Lab (creators of ChatGPT, Character.ai, Mistral, and PyTorch) is hiring a Platform Software Engineer to own core infrastructure systems for Tinker, their fine-tuning API that lets researchers customize frontier AI models. You'll design the authorization layer, build billing and usage metering end-to-end, manage organizations/teams/SSO, implement compliance pipelines, and own audit logging—work that touches nearly every new feature and enterprise deal. Must have a bachelor's in CS or equivalent, backend proficiency in Python or Rust, and demonstrated experience in at least one of: billing/payments, identity/access control, or multi-tenant systems; 4+ years building production backends is preferred, especially with billing at scale, enterprise-readiness patterns, or event-driven metering. Based in San Francisco or New York with $350k–$475k salary and visa sponsorship. The deal-breaker is billing expertise—this role is heavily payments and financial systems focused, requiring strong opinions on idempotency and reconciliation.