Member of Technical Staff - Inference Systems

About the role

Liquid AI is a cutting-edge AI company spun out of MIT CSAIL. We develop efficient AI systems that operate across various deployment targets, from data center accelerators to on-device hardware. Our mission is to ensure low latency, minimal memory usage, privacy, and reliability. We collaborate with enterprises in sectors such as consumer electronics, automotive, life sciences, and financial services. We are currently scaling rapidly and seek talented individuals to join our team.

Responsibilities

• Design and build benchmark suites covering inference performance, model quality, and knowledge evaluation across different hardware targets.
• Run external partner verifications, evaluating their solutions against our benchmarks, identifying gaps, and delivering clear findings.
• Port models like LFM2 onto different runtimes and frameworks, verifying correctness end-to-end.
• Maintain and extend the inference engine layer built on llama.cpp, ONNX, and MLX as new model architectures emerge from research.
• Maintain and extend the inference engine layer built on llama.cpp, ONNX, and MLX as new model architectures emerge from research.
• Maintain and extend the inference engine layer built on llama.cpp, ONNX, and MLX as new model architectures emerge from research.
• Maintain and extend the inference engine layer built on llama.cpp, ONNX, and MLX as new model architectures emerge from research.

Requirements

• Hands-on experience with at least one inference framework like llama.cpp, ONNX Runtime, or MLX, going beyond basic usage into internals and modification.
• Experience designing and building benchmarking pipelines, including methodology, validation, and reproducibility.
• Strong C++ and Python in performance-sensitive contexts.
• Solid understanding of inference fundamentals: quantization, decoding strategies, memory layout, and how they interact.

Qualifications

• Must-have: Hands-on experience with at least one inference framework like llama.cpp, ONNX Runtime, or MLX, going beyond basic usage into internals and modification.
• Must-have: Experience designing and building benchmarking pipelines, including methodology, validation, and reproducibility.
• Must-have: Strong C++ and Python in performance-sensitive contexts.
• Must-have: Solid understanding of inference fundamentals: quantization, decoding strategies, memory layout, and how they interact.
• Desired: Experience porting models across runtimes and verifying numerical correctness.
• Desired: Prior work with external partners or clients in a technical validation or evaluation capacity.
• Desired: Familiarity with edge inference targets and the constraints that come with them.

Skills

• Strong problem-solving skills.
• Excellent communication and collaboration abilities.
• Ability to work independently and as part of a team.

Benefits

• Competitive base salary with equity in a unicorn-stage company.
• Health benefits: We pay 100% of medical, dental, and vision premiums for employees and dependents.
• Financial benefits: 401(k) matching up to 4% of base pay.
• Flexible work schedule: Unlimited PTO plus company-wide Refill Days throughout the year.

Pay

Competitive base salary with equity in a unicorn-stage company.

Schedule

Hybrid employment type.