Responsibilities
THE ROLE:
We are hiring AI / ML Platform Engineers to build the platform layer that makes AI-for-engineering workflows scalable, reliable, and reproducible. This role focuses on the infrastructure and platform systems that support large-scale agent execution, distributed training and inference, experiment tracking, benchmark automation, artifact management, and GPU cluster utilization.
You will work closely with ML Systems Research Engineers, AI Research Scientists, Applied AI Engineers, and hardware domain experts to operationalize the Blueprint framework across kernel optimization, RTL/PPA optimization, ECO fixing, verification, simulation, and debugging workflows.
This is a platform engineering role, not a pure research role. The focus is to build robust shared systems that allow researchers and engineers to run more experiments, compare results reliably, reduce manual orchestration, and move successful workflows into production engineering use.
THE PERSON:
You are a strong systems engineer who enjoys building reliable platforms for AI researchers and applied engineers. You understand distributed systems, ML workloads, GPU infrastructure, experiment management, and production reliability. You can turn messy research workflows into reusable services, APIs, dashboards, job systems, and automation.
You care about reproducibility, observability, performance, and developer experience. You are comfortable working across ML, infrastructure, and hardware/software tooling, and you can partner with research teams without requiring every requirement to be fully specified upfront.
KEY RESPONSIBILITIES:
- Build and operate the shared AI platform for agentic engineering workflows, including job submission, scheduling, orchestration, retries, logging, artifact storage, and experiment tracking.
- Develop reliable infrastructure for distributed training, distributed inference, batch evaluation, and large-scale agent rollout across GPU clusters.
- Build platform services for benchmark execution, correctness checking, profiling, regression tracking, and reproducible evaluation.
- Maintain artifact systems for generated kernels, RTL edits, traces, logs, profiler outputs, benchmark results, simulator outputs, and formal verification artifacts.
- Support scalable integrations with compilers, ROCm/HIP tooling, profilers, simulators, EDA tools, vLLM, SGLang, and internal engineering systems.
- Improve GPU cluster utilization, scheduling efficiency, reliability, quota management, and workload isolation.
- Build dashboards and observability systems for experiment status, resource usage, failure modes, benchmark trends, regression detection, and team productivity.
- Partner with ML Systems Research Engineers to productionize research workflows for RL systems, inference systems, quantification systems, and evaluation pipelines.
- Partner with Applied AI Engineers to make Blueprint harnesses reusable across kernel optimization, RTL optimization, verification, firmware, and CPU/GPU performance workflows.
- Establish platform standards for reproducibility, data retention, run metadata, artifact lineage, access control, and operational reliability.
- TECHNICAL FOCUS AREAS: Distributed ML platform infrastructure for training, inference, evaluation, and agent execution.
- GPU cluster scheduling, utilization, reliability, quota management, and multi-user workload isolation.
- Experiment tracking, artifact management, run lineage, dashboards, and reproducible workflow management.
- Benchmark and evaluation automation for correctness, performance, regression detection, and reproducibility.
- Integration with ROCm/HIP, Triton, compilers, profilers, vLLM, SGLang, simulators, formal tools, and EDA flows.
- Caching and parallelization for expensive feedback loops, including simulator, compiler, verifier, and benchmark workloads.
- Production-quality APIs, services, workflow engines, and developer tooling for research and engineering teams.
PREFERRED QUALIFICATIONS:
- Strong programming skills in Python and one or more systems languages such as C++, Go, or Rust.
- Experience building ML platforms, AI infrastructure, distributed systems, workflow orchestration, experiment platforms, GPU cluster infrastructure, or developer platforms.
- Strong understanding of job scheduling, distributed workloads, logging, monitoring, reliability, storage systems, and production operations.
- Experience with Kubernetes, Ray, Slurm, workflow engines, containerization, CI/CD, data pipelines, or large-scale compute orchestration.
- Ability to build reliable services, APIs, dashboards, and developer tools used by researchers and engineers.
- Strong debugging skills across distributed systems, GPU workloads, storage systems, networking, containers, and production infrastructure.
- Good collaboration skills with AI researchers, ML systems researchers, applied engineers and hardware domain experts.
- PREFERRED EXPERIENCE: Experience with GPU platforms, ROCm/HIP, CUDA, profiling, kernel benchmarking, model serving, or distributed training/inference.
- Experience with vLLM, SGLang, Triton, PyTorch, JAX, Ray, Kubernetes, Slurm, MLflow, Weights & Biases, or similar systems.
- Experience building experiment tracking systems, artifact stores, workflow engines, benchmark automation platforms, or developer productivity tools.
- Experience supporting LLM agents, tool-use systems, large-scale sampling, or automated program optimization workflows.
- Familiarity with compiler, profiler, simulator, formal verification, EDA, firmware, or hardware performance workflows is a strong plus.
- Experience operating shared GPU clusters or high-performance ML infrastructure in a multi-user research environment.
- EDUCATION
- Bachelor's degree in Computer Science, Computer Engineering, Electrical Engineering, Machine Learning, or related field, or equivalent practical experience. Master's preferred; PhD is a plus, especially with work in ML systems, reinforcement learning, distributed systems, GPU computing, or AI infrastructure.
- LOCATION
- : Santa Clara, CA
Qualifications
