Member of Technical Staff (AI Inference Engineer)

Highlights

• Power the inference engine behind every Perplexity query at massive scale.
• Bring new retrieval, text-generation and multimodal models into production.
• Profile bottlenecks and harden reliability across the full serving path.

What they offer

• Offers Equity

Ideal candidate

Self-directed systems engineer who owns problems end-to-end and thrives in fast-moving environments, from research papers to production incidents, with a passion for performance and reliability at scale.

Requirements

• 3+ years professional software engineering experience with ML inference or high-performance systems
• Deep experience with GPU programming and performance work (CUDA, Triton, CUTLASS, or similar)
• Understanding of modern LLM architectures and inference optimization techniques including quantization and speculative decoding
• Experience building and operating production distributed systems under real load, ideally performance-critical
• Comfortable working across Rust, Python, and CUDA/CuTe DSL from runtime to kernels
• Familiarity with at least one deep learning framework (PyTorch, JAX, TensorFlow)
• Understanding of GPU architectures including memory hierarchy, warp scheduling, and tensor cores