GPU Performance tooling engineer

As a GPU Performance Tooling Engineer at Rivos, your responsibilities will include developing and modifying the Open Source Perfetto framework, which is crucial for improving performance metrics of GPU applications. You'll work on underlying libraries to enable performance data collection, ensure minimal monitoring overhead, and build tooling that facilitates performance measurements across different hardware scenarios. Additionally, you'll be responsible for writing unit tests and benchmark tools to validate performance changes, while also keeping abreast of advancements in the field.

To qualify for the GPU Performance Tooling Engineer position at Rivos, candidates should possess a Bachelor’s, Master’s, or PhD in Computer Engineering, Software Engineering, or Computer Science. This educational background provides the necessary foundation in programming, computer architecture, and systems design, all critical for success in this role.

Candidates applying for the GPU Performance Tooling Engineer position at Rivos should be proficient in C or C++, as these languages are essential for developing performance tools. Experience with Protobuf is also required, while familiarity with Rust is a bonus. Understanding deep learning frameworks like TensorFlow or PyTorch can provide additional advantages.

For the GPU Performance Tooling Engineer position at Rivos, soft skills such as problem-solving, written and verbal communication, and teamwork are incredibly important. The role requires the ability to collaborate in a fast-paced environment, manage priorities effectively, and communicate concepts clearly to both technical and non-technical team members.

While not mandatory, having experience with profiling and optimizing low-level performance on GPU architectures can significantly enhance your application for the GPU Performance Tooling Engineer role at Rivos. Additional familiarity with machine learning algorithms or coursework related to them also stands out. A demonstrated passion for open-source contributions would further strengthen your candidacy.

When asked about your experience with Perfetto, highlight specific projects where you've utilized the framework. Discuss the challenges you faced and how you addressed them, emphasizing your ability to contribute to performance optimization through metrics provided by Perfetto.

To address this question, outline various strategies you've implemented in past projects, such as selectively sampling metrics or using efficient data structures. Be specific about how these techniques positively impacted overall system performance.

Discuss your coding standards, practices like code reviews, unit tests, and the use of benchmarking tools. You might also mention how you leverage libraries and third-party tools to maintain code quality and performance.

Use the STAR method to discuss a real-life scenario where collaboration was key. Describe your role, how you communicated ideas, and what the outcomes were, highlighting any performance improvements achieved through teamwork.

Explain how you engage with the broader tech community, through following relevant publications, participating in forums, or attending conferences. Point out any specific resources you find particularly valuable.

Discuss your experience with various hardware platforms and how you adapt your methods for profiling performance. Include any challenges you've faced and how you overcame them to ensure accurate measurements.

Articulate key concepts such as memory hierarchy, cache coherence, and parallel processing. Relate them to how they impact performance and how understanding these concepts has informed your engineering decisions.

Share a systematic approach that includes profiling the model for bottlenecks, adjusting hyperparameters, optimizing data pipelines, and possibly changing the architecture based on insights derived from profiling tools.

Use this moment to explain a specific project, detailing how you managed deadlines, communicated with stakeholders, and ensured deliverables were met without compromising quality or performance.

Express your passion for collaboration and knowledge sharing in the Open Source community. Highlight any previous contributions you've made and how they have enriched your professional journey and technical skills.