Member of Technical Staff

The role

You will work on agent orchestration at scale, environment and task design, and product surfaces around the platform. You will build infrastructure for frontier AI research including dispatch layers, evaluations, and dashboards.

What you will do

• Own the dispatch layer including SQS, concurrency control, and failure handling.
• Build new evaluations and design tasks to test agents.
• Build product surfaces including dashboards for run inspection, tooling for experts, and APIs.
• Support new environment modalities and ship integrations with external orchestration frameworks.
• Monitor model behavior and pod health.
• Debug token throughput and network throughput.

Required qualifications

• Strong engineer across the stack including backend services, data pipelines, and frontend.
• Depth in systems.
• Comfortable running Kubernetes or similar in production.
• Knowledge of auto-scaling, pod lifecycle, persistent storage, and networking.
• Experience building or maintaining message-driven architectures like SQS or Kafka.
• Experience running LLM workloads at scale.
• Knowledge of token instrumentation, rate limit handling, prompt caching, and multi-provider routing.
• Experience building plumbing between models and external tools.

Nice to have

• Ideally at least 3 years at this level

About the company

Chakra Labs builds high-fidelity environments, evals, and datasets for frontier AI research. The team consists of ex-Stripe, Snap, AWS, Microsoft, and Airtable engineers.