Anyscale

We provide a fully-managed compute platform for Ray, an open-source framework developed at UC Berkeley RISELab between 2016 and 2017. Founded in 2019, we help Python developers build and run all their AI applications —from data preparation to training and inference —at any scale. Our platform supports all ML and AI workloads on any cloud and on-premise infrastructure. We also offer a first-party service on Azure Cloud, co-engineered with Microsoft.

Our platform’s key features include:

• Developer Agility: We offer cloud-based, scalable development environments with built-in IDE access (VSCode, Jupyter, Cursor), observability for distributed workloads with profiling tools, and automatic dependency management for Ray nodes.
• Production Resilience: We deploy fault-tolerant Ray clusters on heterogeneous (CPU/GPU) VMs or Kubernetes. We proactively drain and replace unhealthy nodes, offer zero-downtime upgrades, and use managed Prometheus and Grafana dashboards for monitoring and alerting.
• Cost Efficiency: Our Anyscale Runtime optimizes AI pipelines with proprietary technology, manages reliable spot instances (falling back to on-demand when needed), and provides cost governance tools for monitoring usage and managing budgets and quotas.

Our platform supports various AI workloads such as RAG, Stable Diffusion, Batch Inference, Fine-tuning, LLMs, XGBoost, Distributed Training, Video Processing, Generative AI, Data Processing, and Parallel Processing. We also offer multi-tiered support (Developer, Enterprise, Platinum) with different SLAs and features, including email, chat, and dedicated technical resources for higher tiers.

Our mission is to make scalable computing effortless. Our vision is to build the future of distributed computing for AI and ML workflows.

Our core values include:

• Customer & Community Obsessed: We center all decisions on the success of our customers and the Ray community.
• Raise the Bar: We consistently challenge ourselves to improve, think bigger, and push boundaries.
• Take the Longview: We build for the future and remain committed to our vision through challenges.
• LFG: We maintain an ‘LFG’ mindset, focusing on moving fast, pushing boundaries, and celebrating progress.

We operate with a team of 201-500 employees across offices in San Francisco (HQ), Palo Alto, and Bangalore. We invest in our people, reflected in our 4.7 Glassdoor rating and 94% employee recommendation rate. Our work environment emphasizes continuously iterating processes to prevent recurring issues and evolve procedures for company growth. The team that developed the open-source Ray framework is part of Anyscale. They work hands-on with customers to troubleshoot, tune, and scale Ray-based platforms.

We offer a range of employee benefits and perks, including:

• Health, Vision, and Dental Care: We offer a selection of coverage options for employees and their families, with many plans 95% covered.
• Fertility Benefits & Paid Parental Leave: We provide access to fertility benefits through Kindbody (up to $12,500) and 12 weeks of paid leave for new parents.
• Flexible Time Off: Our flexible time off policy allows team members to rest and recharge as needed.
• Learning & Wellness Stipend: A $100 monthly stipend for personal or professional development.
• In-Office Meals: We serve lunch daily in our San Francisco and Palo Alto offices. Dinner is available for those working late, along with boba runs.
• Commute Reimbursement: We cover $15 per day for commuting costs for Bay Area residents.
• Paid Volunteer Time Off: We offer dedicated time off for volunteer work.
• Mental Health Support: We provide mental health support through Rula, which helps find in-network therapists.