Senior AI Scientist

Cerence

4d•$123,500 - $197,600•Hybrid

About The Position

Cerence Inc. 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. At Cerence AI, we help the world's leading automotive and technology brands leverage AI to create safer, more productive and more joy-filled brand and user experiences. We're looking for motivated, collaborative individuals who come alive with big challenges and are excited about AI's potential to shape the future of how people experience the world.

Requirements

Senior AI Scientist
Deep Learning & Transformer Foundations
Optimisation Dynamics & Training Stability
Scaling Laws & Compute Tradeoffs
Loss Functions & Alignment
Distributed Foundation Model Training
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.

Responsibilities

Design and train large-scale transformer and hybrid foundation models
Own model architecture choices across text, multimodal, and emerging paradigms
Diagnose and resolve training instabilities at scale
Navigate scaling tradeoffs across data and compute
Define the technical direction for next-generation models
Apply strong fundamentals in deep learning and representation learning
Design and modify transformer architectures, including: Attention variants, RoPE, ALiBi, Grouped Query Attention (GQA), Mixture-of-Experts (MoE)
Build models from first principles, not just adapt pre-existing codebases
Own optimiser and scheduler choices, including: AdamW, Lion, Adafactor, Learning-rate and warmup schedulers
Understand and debug: Optimiser instability, Gradient pathologies
Apply and validate scaling laws
Navigate Chinchilla-style compute vs data tradeoffs
Make informed decisions about model size, dataset size, and training duration
Design and experiment with loss functions including: Next-token prediction, Contrastive objectives, RLHF, DPO, GRPO
Understand how loss design impacts convergence, generalization, and alignment
Design and execute large-scale training using: FSDP, ZeRO-3, Tensor parallelism, Pipeline parallelism
Apply: Mixed precision (bf16, fp8), Gradient checkpointing
Partner closely with ML systems teams