ML Complier Engineer
Posted 2 days ago by La Fosse
Negotiable
Outside
Remote
United Kingdom
Summary: The role of Compiler Engineer involves collaborating with an engineering team to develop a production-ready machine learning backend for Optical Tensor Processing Units. The position requires extensive experience in compiler construction, runtime systems, and C/C++ programming, along with a strong understanding of machine learning frameworks. The successful candidate will design and build compilers, implement debugging tools, and develop functional simulators. This is a contract position with a focus on delivering high-performance solutions in a fast-paced environment.
Key Responsibilities:
- Design and build a compiler extensible to future optimisation patterns such as op fusion, layout optimisation, tiling, and scheduling.
- Implement optional runtime assertions, compile-time dump mechanisms, TensorBoard timelines, and JAX I/O callback support.
- Develop a simulator that mimics the kernel-space driver, enabling the software team to lower operations ahead of hardware availability.
Key Skills:
- 5+ years of professional experience in C/C++ software engineering.
- Strong background in compilers, runtime systems, and low-level optimisations.
- Deep familiarity with ML frameworks (e.g., PyTorch, JAX, TensorFlow) and their execution models.
- Experience with high-performance computing or hardware-software co-design.
- Strong problem-solving skills with a creative and pragmatic mindset.
- Comfortable operating in fast-paced, ambiguous environments.
- Bachelor’s degree (or higher) in Computer Science, Electrical Engineering, Mathematics, or a related field.
- Excellent communication and collaboration skills.
Salary (Rate): undetermined
City: undetermined
Country: United Kingdom
Working Arrangements: remote
IR35 Status: outside IR35
Seniority Level: undetermined
Industry: IT
Job Description: Compiler Engineer [Contractor]
Company Overview designs and manufactures optical processors built to train and run inference on large-scale AI models.
Contract Overview We are seeking an experienced and highly motivated Compiler Engineer (Contractor) . You’ll collaborate with our in-house engineering team to deliver a production-ready ML backend —including compiler, runtime, debugger, and a functional simulator—for our next-generation Optical Tensor Processing Units (OTPUs). This role requires deep technical expertise in compiler construction, runtime systems, C/C++ programming, and ML framework internals.
Key Responsibilities
- Compiler Development
- Design and build a compiler extensible to future optimisation patterns such as op fusion, layout optimisation, tiling, and scheduling.
- Debugger & Diagnostics
- Implement optional runtime assertions, compile-time dump mechanisms, TensorBoard timelines, and JAX I/O callback support.
- Functional Simulator
- Develop a simulator that mimics our kernel-space driver, enabling the software team to lower operations ahead of hardware availability.
Required Skills & Experience
- 5+ years of professional experience in C/C++ software engineering
- Strong background in compilers , runtime systems , and low-level optimisations
- Deep familiarity with ML frameworks (e.g., PyTorch, JAX, TensorFlow) and their execution models
- Experience with high-performance computing or hardware-software co-design
- Strong problem-solving skills with a creative and pragmatic mindset
- Comfortable operating in fast-paced , ambiguous environments
- Bachelor’s degree (or higher) in Computer Science, Electrical Engineering, Mathematics, or a related field
- Excellent communication and collaboration skills
Contract Details
- Length : 6+ months
- IR35 Status : Outside IR35
- Location : Remote (UK or EU preferred)
- Start Date : ASAP