Job description

A Moving Experience. What You Will Work On Design and operate distributed training systems for large neural networks (autoregressive, diffusion , State Space Models etc. ) across GPU clusters Optimise multi ‑ node, multi ‑ GPU execution to maximize throughput and utilization Diagnose & resolve bottlenecks across compute, memory, and netwo rk Improve training stability and fault tolerance at scale Partner with research and applied ML teams to productioni z e large ‑ model training pipelines Core Responsibilities Distributed Training Infrastructure Build and optimize GPU cluster orchestration using: Slurm Kubernetes Ray RunAI Ensure efficient scheduling, isolation, and fairness across training workloads Communication & Networking Optimize and debug distributed communication using: NCCL RDMA InfiniBand NVLink Minimize networking bottlenecks that dominate end ‑ to ‑ end training time Training Frameworks Scale large-model training using: PyTorch Distributed Megatron ‑ LM DeepSpeed Own multi ‑ node launch configurations, failure recovery, and performance tuning Memory & Performance Optimization Apply advanced memory optimization techniques: Activation checkpointing ZeRO (Stage 1–3) and offload strategies Balance compute, memory, and communication to push model size and batch scale What Success Looks Like GPU utilization consistently stays high (>80–90%) Training scales cleanly from single node to dozens or hundreds of GPUs Communication overhead is minimized and predictable Large training jobs run stably for days or weeks without failure New models can be trained faster, larger, and more reliably than before Required Experience & Skills Strongly Required Deep hands ‑ on experience with distributed systems or ML systems Experience running large ‑ scale workloads on GPU clusters Production experience with PyTorch distributed training Strong understanding of parallelism strategies (data, tensor, pipeline parallelism) Low ‑ level understanding of GPU communication and networking Critical Technical Skills GPU orchestration: Slurm , Kubernetes, Ray, RunAI Communication libraries: NCCL, RDMA, InfiniBand, NVLink Training frameworks: PyTorch Distributed, Megatron ‑ LM , DeepSpeed Memory optimi s ation : activation checkpointing, ZeRO offload techniques Common Problems You’ll Be Solving Many teams fail at scale because: GPU utilization is low despite large clusters Networking and communication dominate training time Training jobs crash or become unstable at large scale You will be explicitly focused on eliminating these failure modes. Ideal Background This role is a strong fit for individuals who have worked as: ML Systems Engineer Distributed Systems Engineer AI Infrastructure Engineer HPC Engineer transitioning into ML Experience working with large language models or foundation models is a strong plus, but deep systems expertise is valued over pure model architecture experience. Why This Role Matters Without robust distributed training infrastructure, progress on large models stalls. This role directly enables: Larger models Faster iteration cycles More reliable research-to-production pipelines You will be building the foundation that makes large ‑ scale AI possible. Cerence Inc. (Nasdaq: CRNC and www.cerence.com ) is the global industry leader in creating unique, moving experiences for the automotive world. Spun out from Nuance in October 2019, Cerence is a new, independent company that has quickly gained traction as a leader in the automotive voice assistant space, working with all of the world’s leading automakers – from Ford and Fiat Chrysler to Daimler, Audi and BMW to Geely and SAIC – to transform how a car feels, responds and learns. Its track record is built on more than 20 years of industry experience and leadership and more than 500 million cars on the road today across more than 70 languages. As Cerence looks to the future and continues an ambitious growth agenda, we need someone to join the team and help build the future of voice and AI in cars. This is an exciting opportunity to join Cerence’s passionate, dedicated, global team and be a part of meaningful innovation in a rapidly growing industry. EQUAL OPPORTUNITY EMPLOYER Cerence is firmly committed to Equal Employment Opportunity (EEO) and to compliance with all federal, state and local laws that prohibit employment discrimination on the basis of age, race, color, gender, gender identity, gender expression, sex, sex stereotyping, pregnancy, national origin, ancestry, religion, physical or mental disability, medical condition, marital status, citizenship status, sexual orientation, protected military or veteran status, genetic information and other protected classifications. Cerence Equal Employment Opportunity Policy Statement. All prospective and current Employees need to remain vigilant when it comes to executing security policies in the workplace. This includes: - Following workplace security protocols and training programs to familiarize with the ways to maintain a safe workplace. - Following security procedures to report any suspicious activity. - Having respect for corporate security procedures to allow those procedures to be effective. - Adhering to company s compliance and regulations. - Encouraging to follow a zero tolerance for workplace violence. - Basic knowledge of information security and data privacy requirements (e.g., how to protect data & how to be handling this data). - Demonstrative knowledge of information security through internal training programs.

Requirements

Must-haves

A Moving Experience. What You Will Work On
Design and operate distributed training systems for large neural networks (autoregressive, diffusion , State Space Models etc. ) across GPU clusters
Optimise multi ‑ node, multi ‑ GPU execution to maximize throughput and utilization
Diagnose & resolve bottlenecks across compute, memory, and netwo rk
Improve training stability and fault tolerance at scale
Partner with research and applied ML teams to productioni z e large ‑ model training pipelines
Distributed Training Infrastructure
Build and optimize GPU cluster orchestration using:
Ensure efficient scheduling, isolation, and fairness across training workloads
Communication & Networking
Optimize and debug distributed communication using:
Minimize networking bottlenecks that dominate end ‑ to ‑ end training time
Training Frameworks
PyTorch Distributed
Megatron ‑ LM
Own multi ‑ node launch configurations, failure recovery, and performance tuning
Memory & Performance Optimization
Activation checkpointing
ZeRO (Stage 1–3) and offload strategies
Balance compute, memory, and communication to push model size and batch scale
What Success Looks Like
GPU utilization consistently stays high (>80–90%)
Training scales cleanly from single node to dozens or hundreds of GPUs
Communication overhead is minimized and predictable

Nice-to-haves

None flagged.

Application prep

Use the role’s language and application details while tailoring your materials.

Keywords to mirror

mlkubernetesai

Platform: workday
Effort: long