Distributed LLM Inference Engineer, Anyscale
$170.1-247k
+ Stock Options
More information about location
3 days a week in office
Scalable compute for AI and Python
Open for applications
Scalable compute for AI and Python
501-1000 employees
Open for applications
$170.1-247k
+ Stock Options
More information about location
3 days a week in office
501-1000 employees
Company mission
To make scalable computing effortless.
Role
Who you are
- Familiarity with running ML inference at large scale with high throughput
- Familiarity with deep learning and deep learning frameworks (e.g. PyTorch)
- Solid understanding of distributed systems, ML inference challenges
Desirable
- ML Systems knowledge
- Experience using Ray Data
- Work closely with community on LLM engines like vLLM, TensorRT-LLM
- Contributions to deep learning frameworks (PyTorch, TensorFlow)
- Contributions to deep learning compilers (Triton, TVM, MLIR)
- Prior experience working on GPUs / CUDA
What the job involves
- As a Distributed LLM Inference Engineer, you will help systems and optimizations that push the boundaries of performance for inference at large scale. This is an incredibly critical role to Anyscale as it allows us to provide market leading performance and price point for AI infrastructure
- Iterate very quickly with product teams to ship the end to end solutions for Batch and Online inference at high scale which will be used by Customers of Anyscale
- Work across the stack integrating Ray Data and LLM engine providing optimizations achieving low cost solutions for large scale ML inference
- Integrate with Open source software like VLLM, work closely with the community to adopt these techniques in Anyscale solutions, and also contribute improvements to open source
- Follow the latest state-of-the-art in the open source and the research community, implementing and extending best practices
Our take
Anyscale operates at the heart of the rapidly expanding AI infrastructure market, where organizations increasingly need systems built to handle large-scale, multimodal, and compiled AI workloads. Founded by the original creators of Ray - an open-source distributed computing framework powering companies like Netflix, Uber, and OpenAI - Anyscale was built to overcome the "AI complexity wall" that legacy compute can't handle. And its providing an AI-native platform that can scale seamlessly from as ingle machine to thousands of GPUs.
The company's impact comes from making distributed AI development faster, easier, and far more efficient. By pairing Ray's flexible compute framework with managed clusters, advanced developer tooling, and performance optimization, Anyscale removes the operational burden of scaling AI. Companies like Canva, Recursion, and RunwayML have seen dramatic gains in utilization, speed and cost. So it's fair to say Anyscale has helped teams deliver AI applications faster and with greater reliability.
The company recently strengthened its position through major partnerships with Azure, CoreWeave, and Google Cloud, bringing AI-native compute directly into enterprise environments. New capabilities like RayTurbo, Kubernetes-native deployment, and first-party managed Ray services make it easier than ever to run high-performance AI at scale. With model sizes and multimodal workloads growing rapidly, Anyscale is in a good position to remain a foundational platform for organizations building the next generation of AI systems.
Kirsty
Company Specialist at Welcome to the Jungle
Insights
133% employee growth in 12 months
Company
Funding (last 2 of 4 rounds)
Aug 2022
$99m
SERIES C
Dec 2021
$100m
SERIES C
Total funding: $259.6m
Company benefits
- Health, Vision, and Dental Care: We offer a wide range of health, dental, and vision coverage options for you and your family—including many plans that are 99% covered by Anyscale
- Fertility Benefits & Paid Parental Leave: Anyscalers can access fertility benefits through Kindbody. Team members can take advantage of 12 weeks of paid leave following the arrival of a new little one.
- Flexible Time Off: We believe all team members should be able to rest up and recharge when needed, which is why we’ve embraced a flexible time off policy at Anyscale.
- Learning & Wellness Stipend: We believe in investing in our #1 asset: our people. We offer a monthly learning and wellness stipend so employees can better themselves at work or pursue personal interests.
- In-Office Meals: Lunch is served everyday in our San Francisco and Palo Alto Offices. Dinner, too, if you’re ever working late. And did we mention boba runs?
- Commute Reimbursement: Whether you’re hopping on public transit or driving yourself to work, Anyscale covers a portion of your commuting costs each month.
- Paid Volunteer Time Off: Give back to the causes and communities you love with paid time off specifically for volunteer work.
- Mental Health Benefits: All Anyscalers can take advantage of included mental health benefits through our Headspace offering.
Company HQ
Yerba Buena, San Francisco, CA
Leadership
Phillip Moritz
(Co-Founder & CTO)Phillip graduated from the University of California, Berkeley with a PhD in Computer Science.
Keerti Melkote
(CEO)Prior to joining Anyscale in 2024, Keerti spent 5 years at Hewlett Packard Enterprise at the President of Intelligent Edge. They also co-founded Aruba Networks where they spent nearly 20 years in roles including CTO and CEO.
Jooree Na
(CFO)Prior to joining Anyscale in 2024, Jooree spent over 5 years at Confluent, joining as a Director of Technical Accounting & Reporting before working through different roles and becoming the VP of Finance.
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
View 2 more jobs at Anyscale