Senior ML Engineer (Performance)

Job Description:

Celestial AI is looking for a Senior ML Engineer to be part of our team and play a critical role in our mission to help our customers bring their ML applications to the next generation of performance driven by our revolutionary photonics-electronics technology.

The ML Engineering Team will be responsible for bringing into play their ML, data science, scientific knowledge, and engineering skills to contribute to the development and delivery of our machine learning reference applications, and work together with customers to map their applications to the Celestial AI hardware.

As a Senior ML Engineer you will be working on reference implementations in a variety of fields: Autonomous Vehicles, Medicine/Radiology/Pathology, Recommendation Engines, Smart Surveillance, Search Engines, Smart Factories, etc, and lead all data science aspects of our customer engagements: complex problem solving, building demos and major customer POC projects.

ESSENTIAL DUTIES AND RESPONSIBILITIES:

• Develop benchmarks for hardware architecture performance analysis at multiple abstraction levels
  • Microbenchmarks directed at specific hardware components such as caches and multithreads
  • Operator-level benchmarks such as in PyTorch
  • Model-level benchmarks such as MLPerf
• Analyze model parallelism and distributed performance on parallel architectures
• Build real-time instrumentation for performance monitoring
• Develop tools and scripts to evaluate and analyze performance at scale to identify hotspots and root causes
• Propose solutions and provide guidance on how to address performance issues
• Participate in research and technology innovation to develop more efficient and automated ways to handle ML/DL workloads running on Celestial AI photonics-electronics hardware, at ExaOp-level scale
• Participate in new product/feature definitions
• Work with the System Software Team to fine tune global resource management for model development and model production-level deployment
• Work with the Machine Learning team to achieve maximum optimization of complex models running on Celestial AI hardware

QUALIFICATIONS:

You must be passionate about ML, data science, versatile and flexible, team player, with a strong can-do mentality. Be a people person with great communication, and ability to work in a fast-paced and dynamic working environment.

• Experience as an ML Engineer or data scientist, either in academia or in industry
• Experience writing code in Python and using Python data-science frameworks
• Strong knowledge of CUDA programming and recent GPU architectures
• Experience with parallel or multi-threaded and distributed programming
• Knowledge in C and C++ in large code bases
• Ability to work with external clients together with great communication skills; ability to talk in webinars, conferences, etc.
• Ability to mentor and develop your talented team members
• Ability to quickly learn new technologies, frameworks, and algorithms
• MS/PhD in Mathematics, Physics, Computer Science, or another related field
• Outstanding English, written and verbal


Location: Santa Clara, CA 

For California location:

As an early startup experiencing explosive growth, we offer an extremely attractive total compensation package, inclusive of competitive base salary and a generous grant of our valuable early-stage equity.  The target base salary for this role is approximately $160,000.00 - $185,000.00. The base salary offered may be slightly higher or lower than the target base salary, based on the final scope as determined by the depth of the experience and skills demonstrated by candidate in the interviews.