Member of Technical Staff (AI Inference Engineer)

📋 Description

• We build and run the inference engine behind every Perplexity query and deploy dozens of model
• Responsibilities include supporting transformer-based retrieval, text-generation, and multimodal
• Migrate GPU kernels to CuTe DSL to run on GB200 today and be portable to Vera Rubin racks.
• Develop a Rust-native internal inference server to handle growing traffic and reduce Python-related
• Profile and optimize performance from network ingress to batching and kernel interleaving.
• Build dashboards, alerts, and automated remediation to catch regressions and respond to production

🎯 Requirements

• Deep experience with GPU programming and performance work (CUDA, Triton, CUTLASS, or similar).
• Understanding of modern LLM architectures and production deployment.
• Experience operating production distributed systems under real load.
• Proficiency across languages and layers: Rust for serving runtime, Python for model code
• Ability to read a research paper, implement a kernel, and debug a production incident within a week.
• Self-directed, thrives in fast-moving environments.

🎁 Benefits

• Equity may be part of total compensation.