Profile cover
A

Junior AI Online Feature Freshness Monitoring Developer

Aubrey Okafor

Freelance · Job · Chicago

About the role

Online feature stores serve model inference in real time. When a feature goes stale — when the last successful update is 4 hours ago for a feature expected to refresh every 15 minutes — the model consuming that feature is making predictions from outdated signals. A real-time fraud model consuming a 4-hour-old transaction velocity count is effectively blind to the recent activity that makes a transaction suspicious. Feature freshness monitoring is the observability layer that catches staleness before it silently degrades model performance. This is a freelance role — 20-25 hours per week. Our feature platform serves 40 ML teams with online and offline feature storage for 2,800 active features. We have basic uptime monitoring. What we don't have is per-feature freshness tracking: monitoring each feature's last update time against its expected refresh cadence, alerting on freshness violations, and tracking the downstream ML models that consume stale features so we can assess incident impact. As a junior developer, you'll build the feature freshness monitoring pipeline: tracking last-update timestamps for each feature in our online store, computing freshness violations against expected cadence SLAs, and building the alerting and dashboard that on-call engineers use to identify and resolve freshness incidents.

Responsibilities

  • Build per-feature last-update timestamp tracking across all features in the online feature store
  • Implement freshness violation detection comparing actual update intervals against expected cadence SLAs
  • Develop downstream impact mapping identifying which ML models consume features with active freshness violations
  • Create freshness monitoring dashboard showing current freshness status by feature group and pipeline
  • Build freshness violation alerting with severity proportional to downstream model impact
  • Maintain monitoring pipeline as new features and refresh cadences are added to the feature store

Requirements

  • 1+ year of Python and SQL development experience
  • Basic familiarity with data pipelines, Airflow, or monitoring systems
  • Interest in ML platform engineering, data reliability, or observability
  • Attention to monitoring accuracy and alert threshold calibration
  • Reliable async communication for a freelance engagement

Benefits

  • Fully remote freelance — 20-25 hours per week
  • Flexible schedule
  • Competitive hourly rate
  • Mentorship from senior ML platform engineers
  • Potential for full-time conversion

Keep exploring related roles

Use these topic pages to discover more roles in the same hiring cluster.

Categories

data science & analytics MLOps & AI Infrastructure

Skills

python sql pandas airflow fastapi

Related remote AI jobs

Nearby roles that share similar skills, categories, or hiring intent.

Browse all jobs

Technical Writer

RemoteAI company

Part-time

Senior Product Security Engineer

RemoteAI company

Full-time Senior

AI Resident UK/EMEA

RemoteAI company

Full-time

Software Engineer, Applications (Scientific Development Solutions) (High Seniority)

RemoteAI company

Full-time Senior

PhD Autonomy Engineer Intern - Computer Vision / Deep Learning Summer 2027

RemoteAI company

Full-time Intern

Drone Platform Integration & Network Support Engineer

RemoteAI company

Full-time

Job Type

Freelance

Level

Junior

Language

English

Salary Range

$60k – $80k / year

AI Expertise

data science & analytics MLOps & AI Infrastructure

Skills

python sql pandas airflow fastapi statistics

Ready to apply for this role?

Create a free talent account in under 2 minutes.

  • Apply to verified AI companies
  • Get AI-matched job recommendations
  • Message hiring managers directly
  • Build your public AI talent profile
Create free account & apply Log in

Interested in this role? Create your free profile to apply.

Apply — create free profile