Manager, Training & Inference Platform at Wayve

London, England, United Kingdom

Apply Now

Not SpecifiedCompensation

Senior (5 to 8 years)Experience Level

Full TimeJob Type

UnknownVisa

Artificial Intelligence, Autonomous VehiclesIndustries

Requirements

Proven Leadership: Strong experience (8+ years) in software engineering

Responsibilities

Maintain and enhance the existing training scheduler (fair-share, preemption, checkpoint/restore)
Provide training introspection (W&B integration, MFU metrics) and debug-node tooling for rapid iteration
Deliver and optimize large-scale GPU inference capacity (persistent & burst)
Enhance Flyte-driven smart scheduling, multi-model inference pipelines, and throughput for hundreds of petabytes of labeling workloads
Grow and mentor the team (will grow to 8+ engineers across both)
Define and drive a unified roadmap, balancing near-term demand spikes with long-term platform resilience
Evolve the scheduler with smart-scheduling features across training and inference workloads
Develop advanced analytics and self-service interfaces to empower ML engineers to configure and monitor their inference workloads effectively
Implement observability & alerting to maintain 99%+ uptime for both platforms
Improve efficiency of the platform with intelligent scheduling techniques and automatic cancellation of non convergent training jobs
Partner with SRE to automate scaling, failover, and incident response
Recruit and develop platform engineers with a broad range of experience (junior through staff)
Foster a culture of ownership, cross-team collaboration, and continuous learning

Skills

Machine Learning

GPU

Model Training

Inference

Platform Engineering

Scheduling

ML Engineering

Wayve

Develops autonomous vehicle technology using AI

About Wayve

Wayve.ai develops self-driving technology known as AV2.0, which focuses on creating a smarter and safer approach to autonomous vehicles. Their technology uses embodied AI software that allows vehicles to learn from their experiences and adapt to different environments without needing detailed programming. This method is different from traditional self-driving technologies that often rely on expensive hardware and pre-mapped data. Instead, Wayve.ai employs end-to-end deep learning, making their solution more cost-effective for automakers. The company targets automakers and fleet operators, offering them adaptable and affordable solutions for driving automation. Wayve.ai has already partnered with major retailers in the UK to test its technology in delivery fleets, aiming to enhance mobility and sustainability in the automotive industry.

London, United KingdomHeadquarters

2017Year Founded

$1,272.3MTotal Funding

SERIES_CCompany Stage

Automotive & Transportation, AI & Machine LearningIndustries

201-500Employees