Member of Technical Staff, TPU Performance Engineering

About the Company

Inferact's mission is to grow vLLM as the world's AI inference engine and accelerate AI progress by making inference cheaper and faster. Founded by the creators and core maintainers of vLLM, we sit at the intersection of models and hardware.

Responsibilities

• Build and optimize TPU backends, compiler integrations, runtime paths, and benchmarking infrastructure
• Improve production-relevant model serving on TPU with clear correctness, latency, and throughput benchmarks
• Work at the boundary of inference systems, kernels, compilers, and hardware architecture
• Make TPU support in vLLM usable, fast, benchmarked, and maintainable

Requirements

• Bachelor's degree in computer science, engineering, systems, machine learning, or similar
• Hands-on experience building or optimizing TPU workloads using JAX, XLA, Pallas, or related tooling
• Deep understanding of TPU execution, memory behavior, compilation, and performance constraints
• Experience optimizing ML kernels or inference paths (attention, GEMM, sampling, KV cache, fused kernels)
• Strong performance profiling and benchmarking skills

Preferred Qualifications

• Experience with vLLM, SGLang, TensorRT-LLM, or XLA-based serving
• Familiarity with batching, KV cache, decoding, and serving tradeoffs
• Experience with compiler technologies such as XLA, MLIR, LLVM, or Pallas
• Knowledge of quantization methods (INT8, FP8, mixed precision)

Benefits

• Annual salary range of S$200,000 to S$400,000 plus equity
• Medical, dental, and vision coverage
• Visa sponsorship available on a case-by-case basis