Senior Staff Software Engineer, AI Runtime

Description

Job ID P-1096.

The company runs one of the largest AI platforms in the world. The team believes the next major advances will come from models trained on specific enterprise data, not only from larger general-purpose models.

The product is a serverless service that makes GPU training at any scale straightforward, so companies can build their own models the way leading frontier labs do. Customers train on it alongside the enterprise data already on the platform. Named customers include Rivian in automotive, Robinhood in fintech, and Merck, which builds foundation models for drug discovery. The company's own AI researchers use it to post-train models ranging from hundreds of billions to nearly a trillion parameters.

Making this work takes substantial engineering. The team has published blog posts on keeping GPUs reliable, making training fault-tolerant, and speeding up fine-tuning with fused kernels and smart sequence packing. AI agents now launch experiments too, and on the research team they have raised training jobs 8x.

This Senior Staff Software Engineer role covers these systems, tackling the hardest technical problems and shaping the product's direction.

The company works in data and AI. More than 20,000 organizations worldwide, including adidas, AT&T, Bayer, Block, Mastercard, Rivian, Unilever, and 70% of the Fortune 500, use its platform to build and scale data and AI apps, analytics and agents. It is headquartered in San Francisco, has 30+ offices around the globe, and offers a unified platform with several data and AI products.

Responsibilities

• Lead the architecture of the training product and set long-term technical direction for AI training infrastructure.
• Define the reliability and observability strategy for a fleet that keeps growing in scale and complexity.
• Shape the developer experience for people and agents, so the largest training runs feel as easy as a first experiment.
• Forecast industry trends and choose technical bets that prepare the product for new GPU generations and workload types.
• Personally take on the hardest multi-node problems, including orchestration, GPU scheduling, data loading and checkpointing.
• Raise efficiency across model architectures and GPU accelerators to make training faster and cheaper.
• Work with research, product and customers to decide what to build next and align teams on hard technical calls.
• Uphold engineering excellence, raise the bar, and help hire and grow strong engineers.

Requirements

• 15+ years in large-scale distributed systems, with deep experience in GPU training, high-performance computing or ML systems.
• Hands-on experience training large models with FSDP, DeepSpeed or Megatron, and a working command of parallelism strategies.
• Deep expertise in GPU performance, including accelerator architecture, high-speed interconnects and kernels.
• Solid grasp of training resilience patterns such as checkpointing, failure detection and automatic recovery.
• Experience building and running multi-tenant cloud platforms as products.
• A record of setting technical direction across teams and carrying it through to business impact.
• The ability to bring teams to agreement on difficult trade-offs, and a reputation as a mentor that senior engineers seek out.
• BS in Computer Science or a related field required; MS or PhD preferred.

Conditions

Local pay range: $228,600 - $342,800 USD, the expected range for this non-commissionable role.

Final pay depends on factors such as job-related skills, depth of experience, relevant certifications and training, and specific work location, and the company expects to use the full width of the range.

The total package may also include eligibility for an annual performance bonus and equity, plus the benefits below.

Access to export-controlled technology or source code may be required; the company decides whether to apply for a U.S. government license and may decline an applicant on that basis alone.

Applicants are considered without regard to age, race, religion, disability, sexual orientation or other protected characteristics.

Benefits

• The company offers comprehensive benefits and perks that vary by region, with specific details available for your location.