Research Engineer - LLM Training & Alignment Systems
Huawei Technologies Canada Co., Ltd. Open · verified Oct 22, 2026 At a glance
Mid-level AI Research role at Huawei Technologies Canada Co., Ltd. Kingston · contract · CA$127,000–CA$225,000 base.
aimlroles summary, based on the employer's posting.
What you'll do
- Build infrastructure and tools for the full foundation-model workflow, from preparing data through training, alignment and evaluation
- Create SFT data workflows for curation, duplicate removal and synthetic examples to improve training inputs
- Tune distributed fine-tuning and alignment pipelines, including reward models and preference optimization across different hardware
- Develop benchmarks that track model quality, alignment, robustness and regressions across training rounds
- Work with research, systems and hardware teams on platform integration, then contribute prototypes, papers and patentable ideas
What you bring
- Hands-on work training and aligning LLMs, including SFT, reward modeling, preference learning and reinforcement-learning optimization
- Experience improving distributed training across multiple devices and machines, including mixed hardware environments
- Experience building SFT data pipelines with semantic deduplication and synthetic-data creation
- Experience designing or using LLM evaluation methods, including automated checks, preference assessment and regression analysis
- Proficiency in Python, C/C++ or Go, and ability to turn research concepts into reproducible systems
Who this fits
This mid-level research engineering contract suits someone who can combine LLM training and alignment with distributed-systems implementation. The role is based on-site in Kingston, Canada, for a twelve-month term.
From the employer
Huawei Canada has an immediate 12-month contract opening for a Research Engineer.
About the team:
The Centre for Software Excellence Lab conducts pioneering research in software engineering, focusing on next-generation technologies. This team integrates industry best practices with cutting-edge academic research to address lifecycle software engineering challenges, including foundation model applications, software performance engineering, hyper-cluster programming, next-gen mobile OS, and cloud-native computing. This lab uniquely allows researchers to apply innovations directly to products affecting billions of customers while promoting open-source contributions, publications, conference participation, and collaborations to create a broader impact.
About the job:
Research, prototype, and build core infrastructure, tooling, and platforms to support the full lifecycle of large foundation model development, including data curation, model training, alignment, and evaluation, with a strong focus on scalability, efficiency, and research impact.
Design and implement systems and workflows for SFT data curation, deduplication, and synthetic data generation, enabling high-quality training signals for large language models. Develop and optimize distributed training and alignment pipelines, including supervised fine-tuning, reward modelling, and reinforcement learning–based preference optimization (e.g., PPO, GRPO), across heterogeneous hardware platforms.
Build and evaluate LLM evaluation and benchmarking frameworks to assess model quality, alignment, robustness, and regression across training iterations. Collaborate closely with systems, hardware, and research teams to integrate novel algorithms and software frameworks into in-house platforms, addressing challenges such as performance modelling, resource allocation, scheduling, fault tolerance, and communication efficiency.
Work with leading industry and academic experts worldwide, contribute to impactful research publications, and drive innovation through prototype systems and patentable inventions that advance large-scale model training and serving.
The total target annual compensation (based on 2,080 hours per year) ranges from $127,000 to $225,000 depending on education, experience, and demonstrated expertise.
About the ideal candidate:
Hands-on experience with large language model training and alignment, including supervised fine-tuning (SFT), reward modeling, preference learning, and reinforcement learning–based optimization (e.g., PPO, GRPO), with a solid understanding of stability, scalability, and efficiency trade-offs .
Strong background in large-scale distributed training systems, with experience optimizing performance, resource utilization, and reliability across multi-node, multi-device, and heterogeneous hardware environments (e.g., GPU, NPU). Experience building SFT data pipelines, including semantic deduplication and synthetic data generation, with a clear understanding of how data quality and distribution affect model behavior and alignment.
Experience designing or applying LLM evaluation and benchmarking frameworks, including automated evaluation, preference-based assessment, and regression analysis to measure model quality, alignment, and robustness.
Proficiency in Python, C/C++, or Go, with the ability to translate research ideas into scalable, reproducible prototype systems, and to communicate technical insights effectively across research and engineering teams.
Additional Information:
Huawei Canada is committed to a fair, inclusive, and accessible recruitment process. If you require accommodation during any stage of the hiring process, please let us know and we will work with you to meet your needs.
All applications for this position are reviewed directly by our hiring team, we do not use artificial intelligence tools to screen or select candidates.