Applied Scientist (Edge AI)

TL;DR

Applied Scientist (Edge AI): Developing and optimizing neural network compression for generative AI models to enable deployment on edge devices and cloud infrastructure with an accent on quantization, pruning, and distillation. Focus on reducing model size by 20x to 100x while preserving quality and building reusable production workflows for partner teams.

Location: Cambridge, UK

Company

hirify.global Devices is an inventive research and development company designing high-profile devices such as Kindle, Echo, and Fire TV.

What you will do

• Apply and extend compression recipes including knowledge distillation, structured pruning, and quantization-aware quantization.
• Design healing recipes through fine-tuning and distillation to recover accuracy lost during the compression process.
• Analyze emerging model architectures to design recovery strategies grounded in the model's internal structure.
• Build a library of reusable model entries, reference implementations, and benchmark results for self-service deployment.
• Define datasets, benchmarks, and KPIs to evaluate accuracy, latency, memory, and cost trade-offs.
• Collaborate with platform engineers and hardware architects to optimize models for specific edge and cloud deployment constraints.

Requirements

• Master's degree or PhD in Computer Science, Electrical Engineering, ML, or a related field.
• Proficiency in programming with Python, C++, or Java.
• Proven track record of publications or patents at top-tier peer-reviewed ML conferences or journals.
• Experience in state-of-the-art deep learning architecture design, training, optimization, and model pruning.

Nice to have

• Experience with multimodal and omni models (vision-language, audio-language, etc.).
• Familiarity with mixed-precision training and inference (FP16, BF16, FP8, INT8, INT4).
• Experience with edge deployment, model compilation, or inference optimization.
• Experience profiling and debugging large-scale ML systems.

Culture & Benefits

• Opportunity to put frontier generative AI into the hands of millions of customers at real scale.
• Ability to publish and present research at top ML venues such as NeurIPS, ICLR, and MLSys.
• Work in a small, fast-moving team with a high degree of ownership over project-level delivery.
• Inclusive culture that values passion for discovery, invention, and simplification.