Member of Technical Staff – Inference Systems

About the role

Liquid AI is looking for a Member of Technical Staff to join its Inference Systems team. You will work on the core inference stack that powers our AI models across data‑center accelerators and on‑device hardware, ensuring low latency, minimal memory usage and high reliability.

Key responsibilities

• Design and implement benchmark suites that evaluate inference performance, model quality and knowledge across diverse hardware targets.
• Run external partner verifications, compare their solutions against internal benchmarks and communicate findings.
• Port models such as LFM2 to multiple runtimes and frameworks, validating numerical correctness end‑to‑end.
• Maintain and extend the inference engine built on llama.cpp, ONNX Runtime and MLX as new architectures emerge.
• Make benchmark results reproducible and explainable for internal teams and partners.

Required profile

• Ability to quickly learn unfamiliar tools and assess their value.
• Strong attention to inference details and trade‑offs, with a rigorous approach to validation.
• Commitment to proving model correctness before declaring a port complete.

Required skills

• Hands‑on experience with at least one inference framework (llama.cpp, ONNX Runtime, or MLX) and ability to modify its internals.
• Proven experience building benchmarking pipelines with reproducible methodology.
• Proficiency in C++ and Python for performance‑critical code.
• Deep understanding of quantization, decoding strategies, memory layout and their impact on inference.