Overview
What this challenge is about.
Rebuild a Fair Single-Selfie Avatar Pipeline at NorthRoom. Expert-level challenge in code. Writing production code that solves real engineering problems, ear...
The Brief
What you'll do, and what you'll demonstrate.
Rebuild a single-selfie-to-riggable-avatar pipeline that runs under 90 seconds per image and delivers quality that is honestly measured and not degraded across demographic groups.
This is not a coding exercise. It is the work a software engineer does between a Jira ticket and a merged PR. That distinction matters to every hiring manager who has seen candidates solve LeetCode problems and none who have shipped production code under real constraints.
When you finish, you will have something most graduates do not: a real-world deliverable, verified by Ewance, that you can show to a hiring manager and say "I did this. Here is the proof."
Earning criteria — what you'll demonstrate
- Fit a parametric face-shape model to landmarks and refine it with a photometric loss to reconstruct a head from one image
- Assemble a multi-stage 3D-from-image pipeline that outputs a riggable mesh conforming to a fixed runtime topology and skinning format
- Design and run a stratified evaluation that surfaces, rather than averages away, per-demographic quality gaps
- Reason about and report latency, quality, and fairness trade-offs as an engineering judgment, not just a metric dump
Program Fit
Where this fits in your program.
Sharpens the same skills your degree expects you to demonstrate.
Aligned coursework coming soon.
Skills
Skills you'll demonstrate.
Each one shows up on your verified credential.
- Digital Humans
Apply digital humans to solve real industry problems and demonstrate production-level capability.
- Geometry Processing
Apply geometry processing to solve real industry problems and demonstrate production-level capability.
- 3d Rendering
Apply 3d rendering to solve real industry problems and demonstrate production-level capability.
- Shape Modeling
Apply shape modeling to solve real industry problems and demonstrate production-level capability.
- Python
Write clean, efficient Python for data processing, automation, and backend services.
- Fairness Evaluation
Apply fairness evaluation 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:
3D Computer Vision Engineer
Reconstructing a head from a single image by fitting a parametric shape model under a photometric loss is the daily work of monocular 3D reconstruction; this challenge has you build and validate that exact loop end to end.
This challenge sharpens
- digital-humans
- geometry-processing
- shape-modeling
Avatar Graphics Engineer
Producing riggable heads that plug into a real-time runtime's fixed topology, texture, and skinning format mirrors avatar-platform engineering, where output must be both lifelike and pipeline-compatible under a tight latency budget.
This challenge sharpens
- 3d-rendering
- digital-humans
- python
Responsible AI Engineer
Designing a stratified evaluation that surfaces demographic quality gaps and turning it into a launch recommendation is core responsible-AI practice for any model that processes images of people.
This challenge sharpens
- fairness-evaluation
- python
- digital-humans