The opportunity
Cerebras Systems builds the world's largest AI chip, 56 times larger than GPUs. This architecture allows Cerebras to deliver industry-leading training and inference speeds; over 10 times faster than GPU-based hyperscale cloud inference services.
What you'll do
Own features end to end. Design, implement, test, and operate features for: chat completions, text generation, streaming, model configuration, tool calling, structured outputs, multimodal inputs, and emerging inference capabilities.
Lead multi-component delivery. Write clear design documents, evaluate: technical tradeoffs, break projects into deliverable milestones, and coordinate implementation with a small group of engineers and partner teams.
Deliver consistent API behavior. Implement clear request and response: semantics across GPU prefill, Cerebras decode, and other inference backends. Maintain compatibility with widely adopted inference interfaces and evolve owned APIs through validation, versioning, and deprecation best practices.
Enable new models and capabilities. Integrate foundation models, tokenizers,: chat templates, generation configuration, sampling methods, and model-specific features into the serving platform.
Integrate serving components. Extend inference services and their integration: with frameworks and runtimes such as vLLM, PyTorch, Hugging Face libraries, LLGuidance, and Cerebras runtime components.
Support disaggregated inference. Implement routing, state-transfer: integration, error handling, retries, cancellation, and request lifecycle management to coordinate GPU prefill with Cerebras decode.
What they're looking for
- Deliver measurable performance improvements. Diagnose bottlenecks and improve: time to first token, streaming latency, throughput, and resource efficiency through changes to tokenization, serialization, batching, scheduling, and communication.
- Own correctness and release quality. Build conformance tests,: model-validation suites, numerical comparisons, integration tests, and performance benchmarks that catch regressions in owned features and support safe releases.
- Strengthen production reliability. Improve service metrics, tracing, logging,: alerts, and diagnostic tools. Investigate production incidents, resolve root causes, and follow through on improvements to resilience and incident response.
- Improve customer and developer experience. Partner with product, support, and: solutions teams to understand user needs and make practical tradeoffs. Improve configuration, documentation, examples, and debugging workflows.