Senior ML Compiler Engineer

5V Tech; Ottawa; On-site; CAD 200,000 - 300,000; Full time

Benefits offered by this job

Equity (Stock Options)

Bonus Plan

Health Insurance

Job summary

5V Tech is seeking a Senior ML Compiler Engineer to help scale a pioneering AI semiconductor platform. You will map deep neural networks onto a novel heterogeneous architecture, leveraging LLVM and MLIR to extend compiler infra for custom hardware.

Collaboration with hardware and ML teams will drive co-optimization for silicon constraints in edge devices. You will join a small, highly selective group building technology from first principles to redefine edge ML processing, with equity and strong

Qualifications

• 5+ years of experience in compiler engineering, preferably within ML/deep learning contexts.
• Strong hands-on experience with LLVM and MLIR.
• Experience with FPGA, analog, or neuromorphic architectures is highly valued.
• Proven ability to translate neural network models onto novel hardware targets.
• Desire to work in a fast-paced, mission-driven startup environment.

Responsibilities

• Build and optimize compilers to map deep neural networks onto a novel heterogeneous processing architecture.
• Work with LLVM/MLIR to extend compiler infrastructure for a custom hardware target.
• Collaborate with hardware and ML teams to co-optimize model performance against silicon constraints.
• Apply compiler design principles to non-traditional computer architectures (FPGA, analog, neuromorphic).
• Contribute to a small, highly selective engineering team building category-defining technology from the ground up.

Skills

Compiler engineering

LLVM

MLIR

Team collaboration

Startup adaptability

Tools

LLVM

MLIR

FPGA

Analog hardware

Neuromorphic hardware

Job description

Senior ML Compiler Engineer

• Canada (Hybrid/Remote)
• Full Time Permanent Contract
• Salary: CAD$200,000 - CAD$300,000 DOE
• Total Comp: Equity (Stock Options), Bonus Plan, Health Insurance, and more

A pioneering AI semiconductor company is building a fundamentally new processing architecture designed to deliver dramatically lower power AI inferencing for always on, edge and endpoint devices. Their proprietary approach breaks from conventional digital compute, enabling real time processing of audio, vision, and sensor data at a fraction of the energy cost of traditional processors, unlocking new possibilities across smart devices, robotics, automotive, and wearable technology.

With proven silicon already validated by customers and fresh investment behind them, they are now scaling their platform software team and are looking for a Senior ML Compiler Engineer to help take their technology from validation into high volume production.

Key responsibilities:

• Building and optimizing compilers to map deep neural networks onto a novel, heterogeneous processing architecture
• Taking machine learning models and neural networks and efficiently fitting them onto unconventional compute arraysWorking with LLVM and MLIR to develop and extend compiler infrastructure for a custom hardware target
• Collaborating with hardware and ML teams to co optimize model performance against silicon constraints
• Applying compiler design principles to non traditional computer architectures (FPGA, analog, neuromorphic)
• Contributing to a small, highly selective engineering team building category defining technology from the ground up

Ideal Profile:

• 5+ years experience in compiler engineering, ideally within ML or deep learning contexts
• Strong hands on experience with LLVM and MLIR
• Experience with FPGA, analog, or neuromorphic computer architectures is highly valued
• Proven ability to take neural network models and translate them onto novel hardware targets
• A genuine passion for cutting edge technology and a desire to make a tangible impact within a small, mission driven team
• Comfortable operating in a fast paced startup environment where adaptability and innovation are essential

This is a rare opportunity to join a small, highly selective team working on technology built from first principles to redefine how machine learning is processed at the edge. Full details on the company will be shared once you're through to the next stage.