AI Service Dependency Impact Modeler

Evorra Systems

Remote · $80k – $100k / 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

When Service X degrades, which other services break, and by how much? At most companies, the answer lives in someone's head. At ours, it should be in a model. We've built the incident management platform that tracks what breaks. What we haven't built is the dependency graph that explains why — and more importantly, the impact predictor that tells on-call engineers which downstream systems to check immediately and which can wait. You'll build the service dependency impact model: mining incident and deployment event logs to infer service dependencies automatically, building the impact propagation model that predicts blast radius from a single service failure, and creating the real-time scoring that surfaces impact predictions during active incidents. The output is a ranked list of services that are likely affected by the current incident, updated as new telemetry arrives. The analytics team is four. You'll own this feature from prototype to production. The senior infrastructure architect who understands the service topology will be your main collaborator for validating model outputs against their domain knowledge.

Fully remote.

Responsibilities

Mine incident and deployment logs to infer service dependency relationships
Build impact propagation model predicting blast radius from service failures
Create real-time scoring that updates impact predictions during active incidents
Develop confidence calibration for impact predictions used in high-stakes situations
Build evaluation framework validating predictions against historical incident records
Design APIs integrating impact predictions into the incident management dashboard

Requirements

3+ years in data science or analytics with distributed systems or infrastructure domain experience
Experience with graph analysis and network propagation models
Strong Python and SQL skills for data pipeline development
Understanding of microservice architectures and failure cascade patterns
Ability to work with infrastructure engineers to validate model assumptions