Coframe
Agentic System Engineer
About this role
Join Coframe to engineer AI agent systems that automatically generate, validate, and monitor production-ready web interfaces and features at scale. You'll build robust systems combining LLMs, evaluation pipelines, and continuous monitoring to accelerate experimentation velocity and product optimization.
What you'll do
- Build AI agents that generate interfaces and client-side features at scale
- Create validation systems to catch bugs, brand inconsistencies, and poor user experiences before deployment
- Develop online and offline evaluation systems for generated outputs
- Build monitoring infrastructure to detect issues and regressions post-launch
- Improve orchestration between generation, QA, rollout, and learning loops
- Convert manual workflows into reliable, scalable agent-driven processes
What they're looking for
- Production AI systems and agent workflows
- LLM pipelines and tool use
- Building reliable non-deterministic systems
- Evaluation and monitoring systems
- Web development and product sense
- Parallel execution and orchestration
- Stakeholder communication and requirements translation
- Software engineering fundamentals
Benefits
- Unlimited paid time off
- Medical, dental, and vision coverage for family
- Life insurance and disability benefits
- 401k with company match
- In-office meals provided
- Gym membership
Opens the official application on the employer’s site. No login required.
Coframe
Coframe builds an AI-powered growth platform that automatically generates, validates, and optimizes web interfaces and features at scale using LLMs and continuous monitoring. The company is hiring Solutions Engineers, AI systems engineers, infrastructure engineers, Product Engineers, and Forward Deployed Engineers to implement customer integrations, build core platform systems, and drive enterprise deployments.
- Website
- coframe.ai
Likely interview questions
- Walk us through a production AI system or LLM pipeline you've built—how did you handle non-determinism and ensure reliability at scale?
- Describe your experience building evaluation systems for LLM outputs. How do you balance automated metrics with qualitative quality signals?