At a Glance
- Tasks: Optimise advanced language models and enhance training performance with cutting-edge tools.
- Company: Join a leading tech firm at the forefront of natural language processing.
- Benefits: Competitive salary, flexible work options, and opportunities for professional growth.
- Other info: Collaborate with top researchers and enjoy a dynamic, innovative work environment.
- Why this job: Make a real impact in AI by improving model performance and driving innovation.
- Qualifications: Strong software engineering skills, proficiency in Python, and experience with ML frameworks.
The predicted salary is between 60000 - 80000 € per year.
Requirements
- Extremely strong software engineering skills
- Proficiency in Python and related ML frameworks such as JAX, Pytorch and XLA/MLIR
- Experience writing kernels for GPUs using CUDA, triton, etc
- Experience using large-scale distributed training strategies
- Familiarity with autoregressive sequence models, such as Transformers (Desirable)
- Bonus: paper at top-tier venues (such as NeurIPS, ICML, ICLR, AIStats, MLSys, JMLR, AAAI, Nature, COLING, ACL, EMNLP)
If some of the above doesn’t line up perfectly with your experience, we still encourage you to apply!
What the job involves
- As a Performance Engineer in the Pre-Training team you will be responsible for optimizing the performance of our advanced language models and systems.
- Their primary focus is on improving key model training metrics, such as training throughput, ensuring high accelerator utilization.
- The team combines expertise in software engineering, machine learning, and low-level kernel design and development to design robust systems and enhance model performance.
- You will work on identifying and removing performance bottlenecks, develop cutting‑edge training and profiling tools to help Cohere's mission of providing efficient and reliable language understanding and generation capabilities and drive innovation in the field of natural language processing.
- Design and write high-performant and scalable software for training.
- Understand architectural modifications and design choices and their effects on training throughput and quality.
- Write low-level CUDA, triton kernels to squeeze every last bit of performance from our accelerators.
- Research, implement, and experiment with ideas on our supercompute and data infrastructure.
- Learn from and work with the best researchers in the field.
Member of Technical Staff (Training Performance Engineer) employer: Deepstreamtech
Cohere is an exceptional employer for those passionate about advancing natural language processing, offering a dynamic work culture that fosters innovation and collaboration. Located in a vibrant tech hub, employees benefit from access to cutting-edge resources, opportunities for professional growth, and the chance to work alongside leading experts in the field. With a strong emphasis on employee development and a commitment to pushing the boundaries of technology, Cohere provides a rewarding environment for technical talent to thrive.
StudySmarter Expert Advice🤫
We think this is how you could land Member of Technical Staff (Training Performance Engineer)
✨Tip Number 1
Network like a pro! Reach out to folks in the industry, attend meetups, and connect with people on LinkedIn. You never know who might have the inside scoop on job openings or can refer you directly.
✨Tip Number 2
Show off your skills! Create a portfolio showcasing your projects, especially those involving Python, CUDA, or ML frameworks. This gives potential employers a taste of what you can do and sets you apart from the crowd.
✨Tip Number 3
Prepare for technical interviews by brushing up on your coding skills and understanding the latest in ML and performance engineering. Practice common interview questions and maybe even do mock interviews with friends or mentors.
✨Tip Number 4
Don’t hesitate to apply through our website! Even if you don’t tick every box in the job description, we value diverse experiences and perspectives. Your unique background could be just what we’re looking for!
We think you need these skills to ace Member of Technical Staff (Training Performance Engineer)
Some tips for your application 🫡
Show Off Your Skills:Make sure to highlight your software engineering skills and any experience with Python and ML frameworks like JAX or Pytorch. We want to see what you've got, so don’t hold back!
Tailor Your Application:Customise your application to reflect how your experience aligns with the role. If you’ve worked on GPU kernels or distributed training strategies, let us know! We love seeing relevant examples.
Don’t Sweat the Small Stuff:If you don’t tick every box in the job description, don’t worry! We encourage you to apply anyway. Your unique experiences might just be what we’re looking for.
Apply Through Our Website:For the best chance of getting noticed, make sure to apply through our website. It’s the easiest way for us to keep track of your application and get back to you quickly!
How to prepare for a job interview at Deepstreamtech
✨Show Off Your Coding Skills
Make sure to brush up on your Python and any ML frameworks you’ve worked with, like JAX or PyTorch. Be ready to discuss specific projects where you wrote kernels for GPUs using CUDA or Triton, as this will demonstrate your technical prowess.
✨Know Your Models
Familiarise yourself with autoregressive sequence models, especially Transformers. Be prepared to explain how these models work and how you’ve applied them in past projects. This shows you’re not just a coder but also understand the underlying concepts.
✨Prepare for Performance Questions
Since the role focuses on optimising performance, think about past experiences where you identified and removed performance bottlenecks. Have examples ready that highlight your problem-solving skills and your ability to enhance training throughput.
✨Research the Company and Team
Take some time to learn about the company’s mission and the Pre-Training team’s goals. Understanding their focus on efficient language understanding and generation will help you align your answers with what they value most during the interview.
