Careers
Build the infrastructure for training the next generation of models.
Early members define key systems and key business — not pre-scoped execution. Below is who we are, why to join, and the roles we’re hiring for.
About us
Turning complex work into trainable systems.
Copula Lab is building the data and environment infrastructure for training the next generation of AI models.
As models move from coding and general Q&A into more complex knowledge work, capability gains increasingly depend on high-quality task environments, expert feedback, verifiable benchmarks, and training-feedback loops. Real-world complex tasks can’t be solved by simple labeling — they involve long context, multi-step operations, professional judgment, tool use, multimodal understanding, and result reliability.
Our job is to turn these complex tasks into data and environment systems that models can train, evaluate, and continuously iterate on.
Why join
Define key systems from 0 to 1.
We’re building our core direction, product system, customer network, and organization from 0 to 1 — early members define key systems and key business, not pre-scoped execution.
You’ll work against the frontline needs of leading model teams across post-training, agent training, eval, RL environments, expert data, and training-feedback loops — and go deep into real tasks inside high-value knowledge workflows, turning model failure modes into benchmarks, task environments, data products, and training-feedback plans.
For people who can independently own a key direction and consistently deliver, we offer competitive cash, early-employee equity, and room to grow into a direction lead or executive.
Open roles
Research Scientist
This is not a traditional model-tuning or applied-ML role. We need someone who understands frontier capability iteration — post-training, eval, agent training, and data strategy — and can also analyze how models fail in real business scenarios. Your core output is research judgment, evaluation methods, data strategy, and experimental conclusions. You’ll independently own a research or data direction, turning fuzzy problems into clear, deliverable benchmarks, data products, or training-feedback plans.
View roleEngineeringAgent Training Infra Engineer
This is not a traditional full-stack role. We need someone who can build systems, platforms, and tools, and also understands agent training, model data, and task-environment construction. You don’t need to start as an RL expert, but you should have touched related systems or data pipelines and be willing to go deeper. Your core output is runnable, verifiable, scalable task environments and data-production systems.
View roleProductAgent Data Strategy — Product
This is not a traditional SaaS PM, nor a role that only writes PRDs and draws prototypes. We need someone who understands agent product forms, capability boundaries, and eval methods, and can go deep into real business scenarios to decompose complex tasks with domain experts. Your core output is real-scenario task definitions, expert-collaboration processes, task specs, and deliverable product plans — turning real business tasks into task environments and data products that agents can execute, evaluate, and iterate on.
View roleGo-to-marketBD / Enterprise Sales
This is not a traditional software-sales role, nor a BD role that only cold-calls, books meetings, and follows up on leads. We need someone who understands what model teams really need in post-training data procurement, and can drive procurement to close inside big tech and model teams. Your core output is high-quality customer relationships, clear needs judgment, a convertible pipeline, and final orders — turning model teams’ fuzzy training, eval, and data-procurement needs into products and project plans the company can deliver, accept, and repeatedly repurchase. For people who can consistently open key accounts and land high-quality orders, we offer commercially-tied cash incentives and early-employee equity.
View role