Overview
What this challenge is about.
Prune and distill a speech model for a hearable, then document the accuracy trade-offs to earn your verifiable certificate.
The scenario
The startup (around 60 people, post-Series-A) ships about 200,000 units per year across the Nordics and Germany and needs the always-on KWS to stay within the DSP's free 50 KB SRAM so the rest of the audio stack still runs.
The Brief
What you'll do, and what you'll demonstrate.
Compress a keyword-spotting model into a 50 KB INT8 footprint via pruning + distillation while keeping accuracy production-viable.
Earning criteria — what you'll demonstrate
- Apply structured pruning to a real CNN keyword spotter
- Use knowledge distillation to compress a larger model into a smaller student
- Evaluate KWS models with the metrics that matter for product (FAR at fixed FRR)
- Communicate compression trade-offs to firmware leadership
Program Fit
Where this fits in your program.
Sharpens the same skills your degree expects you to demonstrate.
Edge ML and On-Device Machine Learning
Master · Ai Systems
Strong alignment
This challenge maps to Edge ML and On-Device Machine Learning at the Master level. It sharpens the same practical skills your coursework expects — but in a real industry context with actual constraints and deliverables.
Skills
Skills you'll demonstrate.
Each one shows up on your verified credential.
- Pruning
Apply pruning to solve real industry problems and demonstrate production-level capability.
- Knowledge Distillation
Apply knowledge distillation to solve real industry problems and demonstrate production-level capability.
- Model Compression
Apply model compression to solve real industry problems and demonstrate production-level capability.
- Edge Inference
Apply edge inference to solve real industry problems and demonstrate production-level capability.
- Quantization
Apply quantization to solve real industry problems and demonstrate production-level capability.
- Pytorch
Apply pytorch to solve real industry problems and demonstrate production-level capability.
Careers
Career paths this challenge builds toward
Completing this challenge demonstrates skills that transfer directly to these roles:
Machine Learning Engineer
Hitting a strict on-device SRAM budget through pruning + distillation is the daily job of an MLE on any consumer-audio or hearable team.
This challenge sharpens
- pruning
- knowledge-distillation
- model-compression
MLOps Engineer
Reproducible compression + deployment pipelines for embedded targets transfer directly to MLOps roles at edge-AI companies.
This challenge sharpens
- edge-inference
- quantization
- model-compression
Applied AI Scientist
Mapping a compression budget into an effort/accuracy memo for firmware leadership is the applied-AI-scientist craft of turning research into product decisions.
This challenge sharpens
- knowledge-distillation
- pruning
- pytorch