Note: The job is a remote job and is open to candidates in USA. Stack AV is developing revolutionary AI and advanced autonomous systems designed to enhance safety, reliability, and efficiency in modern operations. In the Staff Software Engineer role, you will define and drive architecture for a high-throughput, low-latency, multi-tenant ML inference platform, balancing hands-on coding with long-term technical direction.

Responsibilities

• Design platform architecture for multi-tenant inference workloads across serving, orchestration, control plane, APIs, SDKs, observability, and model-engine integration
• Develop robust API layers (gRPC, WebSockets, REST, etc.) and developer SDKs that abstract complex distributed inference orchestration into seamless, reliable token streams
• Build and harden a multi-tenant control plane to enable accurate metering, rate limiting, quotas, tenant isolation and noisy-neighbor fairness across the platform
• Optimize inference performance across the entire system stack, including the model engine layer
• Build observability and SLOs to gain insights into system economics, cache-hit rates, GPU utilization and cost accounting per model and per tenant
• Partner with product and infrastructure teams on model onboarding, capacity planning, external API contracts and customer adoption
• Promote Engineering Excellence: Maintain a high bar for engineering excellence in their own work but also set a culture of engineering excellence within the team

Skills

• Education: Bachelor's or Master's degree in Computer Science, Engineering, or a related field
• Experience: 7+ years of experience building and operating backend distributed systems end to end
• Demonstrated cross-team technical leadership in backend distributed systems, ML infrastructure, inference serving, or high-performance compute platforms
• Strong Data & ML systems fundamentals: data-intensive distributed systems, concurrency, networking and performance profiling
• Hands-on experience running large-scale inference services on GPUs, including KV caches, prefill/decode stages and throughput/latency trade-offs
• Direct experience with inference engines (TensorRT, vLLM, etc) or serving frameworks (Dynamo, Triton or equivalent)
• Technical Skills: Strong programming skills in C++, Go, Rust or Python
• Familiarity with deep learning frameworks (PyTorch, etc.) as well as model parallelism
• Familiarity with GPU computing primitives such as CUDA, NCCL, NVLink, and hardware-specific optimizations
• Practical understanding of high-performance networking architectures, including InfiniBand, RoCE, and low-latency cluster communication
• Communication: Excellent verbal and written communication skills, with the ability to convey complex technical concepts to non-technical stakeholders
• Autonomous vehicles (AV) experience is a bonus

Company Overview