Junior AI Model Version Compatibility Verification Developer

Rolvex Solutions

Remote · $60k – $80k / year · Full-time · Posted 4 months ago

This role is no longer accepting applications

Don't miss the next role like this

Create a free profile with your skills. We match you to new AI roles as they open, and companies hiring can contact you directly.

Create free profile, 2 minutes

Similar open roles you can apply to now:

See all open AI jobs

This role is no longer accepting applications

Don't miss the next role like this

Create a free profile with your skills. We match you to new AI roles as they open, and companies hiring can contact you directly.

Create free profile, 2 minutes

Similar open roles you can apply to now:

See all open AI jobs

About the role

Deploying a new model version seems like it should be safe once the model passes accuracy tests. But a model that changed its output schema — returning a different confidence score format, adding a new class label, or dropping an optional field — can break the downstream applications that consume its predictions even when the model itself is more accurate. Compatibility verification catches these breaking changes before deployment reaches production. Our deployment platform manages model deployments for 45 enterprise ML teams. Version compatibility checking is a gap in our pre-deployment gate. We validate model performance against test sets. We don't systematically check whether the new version's output format, latency profile, and behavioral distribution are compatible with the downstream consumers that are registered for that model endpoint. As a junior developer, you'll build the model version compatibility verification components: output schema comparison between versions, behavioral distribution diff that detects meaningful changes in model output distributions, and the compatibility report generation that deployment engineers review before approving a rollout.

Fully remote.

Responsibilities

Build output schema comparison detecting format changes between new and existing model versions
Implement behavioral distribution diff measuring output distribution changes across production traffic samples
Develop downstream consumer registry tracking which applications depend on each model endpoint version
Create compatibility report generation summarizing breaking changes and behavioral differences for deployment review
Build regression baseline comparison checking new model versions against established production behavior
Maintain compatibility checks as new model types and output formats are added to the platform

Requirements

1+ year of Python and SQL development experience
Basic familiarity with REST APIs, data schemas, or software versioning concepts
Interest in ML platform engineering, DevOps, or software reliability
Attention to correctness in schema comparison and behavioral difference detection
Good communication for collaborating with deployment engineers on compatibility requirements