About the Role

LM Studio is the world's most popular desktop application for running AI locally. The performance and flexibility of our inferencing engines is core to what people love about LM Studio and what makes us unique. In this role you will bring up new model support in LM Studio's inferencing libraries as well as in open source engines (e.g. llama.cpp, MLX, and more).

As an On-Device ML Engineer working on LM Studio, you are expected to be able to implement new model code across inferencing libraries, port implementations from (e.g. PyTorch / HF transformers) to other libraries (e.g. llama.cpp), or implement models based on academic papers.

Description

Bring up new AI model architectures across LM Studio's integrated inferencing libraries and ship them to millions of users worldwide. Collaborate with experts in the open-source community to make new model support available as soon as possible. Leverage software engineering skills and excellent intuition to architect and implement extensible on-device ML software. Work with LM Studio systems and app engineers to make the latest AI technologies useful and pliable through UIs and APIs.

Minimum Qualifications

• 2+ years experience implementing ML models using PyTorch, HF transformers, llama.cpp, or similar
• 2+ years in C++ and Python programming
• Excellent communication and collaboration skills
• Disposition towards for local AI and local on-device ML

Preferred Qualifications

• Deep understanding of ML model architectures and algorithms
• Experience methodically evaluating algorithm performance
• Experience contributing to open source projects

Pay and Benefits

• Equity
• Medical, Dental, Vision insurance
• 401K
• Company provided MacBook Pro M3 / or high performance Windows/Linux laptop
• Access to awesome hardware of all kinds
• Office in the Navy Yard, BK with stunning views and free gym access

The expected salary range for this position is:

Annual Salary Range:

$150,000 - $350,000 USD