Huawei Canada has an immediate permanent opening for a Senior Deep Learning
Compiler Engineer.
About the team:
Initially founded in 1991 as Huawei's ASIC Design Center, the IC Lab is a
leading global fabless semiconductor lab. This lab delivers trusted,
cutting-edge semiconductor products and services for smart devices, contributing
to smart home and mobility solutions. The local team in Canada specializes in
semiconductors, and chipset solutions.
About the job:
-
Lead research efforts to develop state-of-the-art algorithms that
significantly improve deep learning model inference times, memory usage, and
overall performance. Focus on ground breaking techniques that push the
boundaries of compiler optimization, and drive innovations that directly
impact the effectiveness of the system in real-world wireless applications.
-
Drive initiatives that accelerate the development workflow within the team.
You will identify bottlenecks, streamline processes, and introduce best
practices to ensure that the team remains agile, efficient, and able to
deliver high-quality results quickly. You’ll play a key role in defining
engineering practices that enhance both individual and team output.
-
Provide mentorship to junior engineers and foster a culture of continuous
learning. You’ll guide their professional development, help refine their
skills, and ensure the team remains technically strong and aligned with best
practices. Additionally, you’ll review designs and code to ensure high
standards are maintained while providing constructive feedback to elevate the
team's overall capabilities.
-
Lead the design and development of optimized code in C++ and Python for deep
learning inference systems. You will take ownership of performance-critical
components and help define appropriate high-level abstractions that balance
clarity with efficiency.
-
Work closely with other engineering teams to ensure that optimizations and
new algorithms are implemented efficiently across the entire system. Your
ability to clearly communicate complex ideas and collaborate with
cross-functional teams will be crucial in driving system-level changes that
result in tangible improvements.
-
Lead efforts to integrate and optimize deep learning systems for custom
hardware accelerators, such as GPUs, TPUs, or other AI accelerators. Your
deep understanding of hardware-software interactions will drive major
performance improvements in real-time applications, ensuring that the system
runs at peak efficiency.
-
Work closely with product managers and other stakeholders to help define the
technical roadmap. You will take an active role in determining the direction
of the technology stack and prioritize initiatives based on business needs
and technological advancements.
About the ideal candidate:
-
5+ years of experience in software engineering or a related domain, with a
proven track record of developing complex systems and optimizing deep
learning pipelines. A background in compiler design or machine learning
systems is an asset.
-
Deep knowledge of C++ and Python, with experience writing optimized,
high-performance code. Strong understanding of algorithms, data structures,
and performance profiling.
-
Passion for researching new algorithms that improve deep learning systems.
You stay up-to-date with the latest research in compiler optimization, deep
learning, and hardware acceleration, and you’re eager to apply new findings
to solve practical challenges.
-
You are an experienced mentor who enjoys guiding junior engineers, sharing
knowledge, and helping others grow technically. Experience reviewing code,
providing feedback, and driving technical discussions to ensure high-quality
engineering standards.
-
Excellent problem solver, with the ability to navigate complex technical
challenges independently. You take initiative, and you excel at driving
projects forward with minimal supervision.
-
Ability to articulate complex technical concepts clearly to both technical
and non-technical stakeholders is critical. You thrive in a collaborative
environment and can effectively balance individual contributions with team
goals.
-
Experience with MLIR (Multi-Level Intermediate Representation) and other
specialized compiler frameworks is an asset, as is a background in optimizing
deep learning models for edge devices and real-time applications.
