Member of Technical Staff, Essential AI
ML Infrastructure, Platform Engineer
$191.3-225k
+ Equity
5 days a week in office
AI products for enterprise
Job no longer available
AI products for enterprise
21-100 employees
Job no longer available
$191.3-225k
+ Equity
5 days a week in office
21-100 employees
Company mission
To build an open platform to fuel and accelerate AI breakthroughs globally.
Role
Who you are
- A strong understanding of architectures of new AI accelerators like TPU, IPU, HPU etc and their tradeoffs
- Knowledge of parallel computing concepts and distributed systems
- Prior experience in performance tuning of training and/or inference LLM workloads. Experience with MLPerf or internal production workloads will be valued
- 6+ years of relevant industry experience in leading the design of large-scale & production ML infra systems
- Experience with training and building large language models using frameworks such as Megatron, DeepSpeed, etc and deployment frameworks like vLLM, TGI, TensorRT-LLM etc
- Comfortable with working under-the-hood with kernel languages like OAI Triton, Pallas and compilers like XLA
- Experience with INT8/FP8 training and inference, quantization and/or distillation
- Knowledge of container technologies like Docker and Kubernetes and cloud platforms like AWS, GCP, etc
- Intermediate fluency with network fundamentals like VPC, Subnets, Routing Tables, Firewalls etc
What the job involves
- The ML Infra Platform Engineer will be responsible for architecting and building the compute infra that powers the training and serving of our models
- This requires a full understanding of the complete backend stack → from frameworks to compilers to runtimes to kernels
- Running and training models at scale often requires solving novel system problems
- As an Infra Systems Engineer, you'll be responsible for identifying these problems and then developing systems that optimize the throughput and robustness of distributed systems
- With proven experience building large-scale platforms, you will be responsible for building and advancing our systems that allow research and engineering organizations to iteratively develop, test, and deploy new features reliably, with high velocity, and with a frictionless-fast development cycle
- Design, build, and maintain scalable machine learning infrastructure to support our model training, inference and applications
- Design and implement scalable machine learning and distributed systems that enable training and scaling of LLMs. Work on parallelism methods improve training of in a fast and reliable way
- You will help oversee and drive the vision of how we should build, test, and deploy models, while taking ownership and transform state-of-the-art development experience for research
- Develop tools and frameworks to automate and streamline ML experimentation and management
- Collaborate with other researchers and product engineers to bring magical product experiences through large language models
- Working on lower levels of the stack to build high-performing and optimal training and serving infrastructure including researching new techniques and writing custom kernels as needed to achieve improvements
- Be willing to optimize performance and efficiency across different accelerators
Our take
Essential AI is on a mission to make frontier AI accessible to everyone. The team builds open models, easy-to-use tools, and reliable pipelines so anyone can experiment, iterate, and create faster. Founded by a team with experience at Google and other tech giants, the company combines deep know-how with a curiosity-driven approach to AI.
Even though it's still early days, Essential AI is making waves by sharing cutting-edge models and open science along the way. This approach doesn't just accelerate research but it invites collaboration and helps new talent jump in as well as turns complex AI breakthrough into real-world tools people can actually use.
With funding and a growing team too, Essential AI is now focused on building products that can really change how businesses work, making workflows smarter and productivity smoother. By partnering with builders and developers around the world, the company is shaping the future of AI in a way that's open, practical, and pretty exciting.
Freddie
Company Specialist at Welcome to the Jungle
Company
Funding (2 rounds)
Dec 2023
$56.5m
SERIES A
Mar 2023
$8.3m
SEED
Total funding: $64.8m
Company HQ
SoMa, San Francisco, CA
Leadership
Ashish Vaswani
(Co-Founder & CEO)Previously co-founder and Chief Scientist of Adept AI Labs and Staff Research Scientist at Google Brain.
Salary benchmarks
We don't have enough data yet to provide salary benchmarks for this role.
Submit your salary to help other candidates with crowdsourced salary estimates.
Share this job