DeveloperJobs.io
← Back to all jobs

Job Description

Join Unity Technologies SF in Mountain View to build and operate machine learning that helps prevent ads/offerwall fraud in real time. This role combines end-to-end ownership of ML models, data labeling and evaluation rigor, and modernization of training and serving infrastructure, with a strong focus on production reliability and monitoring.

What you’ll do

  • Ship fraud models to production: Move real-time anomaly detection and device risk models from development into production so outputs drive payout and blocking decisions.
  • Build and maintain ground truth: Create and keep a golden labeled fraud dataset, define evaluation methodology, and establish baseline fraud-efficacy metrics used to judge model and rule changes.
  • Own features end to end: Design and evolve streaming and batch features feeding a Redis online feature store, partnering with data engineering on freshness, correctness, and lineage.
  • Modernize ML infrastructure: Plan and drive evolution of the training and serving stack, including assessing a potential move from SageMaker to Kubeflow and KServe, with attention to latency for real-time scoring.
  • Detect new fraud patterns: Investigate fraud spikes, analyze user and device behavior, and translate findings into new detection signals, features, and models.
  • Operate and document: Monitor model and feature health in production, write runbooks and data lineage documentation, and participate in on-call for the fraud detection stack.

What you bring

  • Production ML experience: Several years building ML models and taking them through deployment, monitoring, and iteration after launch.
  • Fraud, risk, or anomaly detection: Hands-on experience with fraud detection, abuse prevention, risk scoring, or anomaly detection, including working with sparse, noisy, or adversarial labels.
  • Real-time serving and feature engineering: Low-latency model serving experience plus online/offline feature pipeline experience (for example Redis, Kafka, or Airflow).
  • Evaluation rigor: Ability to build labeled datasets from imperfect signals and design evaluation methods that remain effective as fraud patterns shift.
  • Core tools: Strong Python and SQL, plus familiarity with cloud ML platforms such as AWS SageMaker, Kubeflow, or KServe.
  • Ownership and collaboration: Comfortable taking ambiguous problems to shipped outcomes with limited oversight, documenting work, and partnering with backend, data, and product teams (including using AI coding tools to move faster).

Tools you’ll work with

Python, SQL, Redis, Kafka, Airflow, AWS SageMaker, Kubeflow, KServe, Kubernetes

Benefits at Unity Technologies SF

  • Comprehensive health, life, and disability insurance
  • Commute subsidy
  • Employee stock ownership
  • Competitive retirement/pension plans
  • Generous vacation and personal days
  • Support for new parents through leave and family-care programs
  • Office food snacks
  • Mental Health and Wellbeing programs and support
  • Employee Resource Groups
  • Global Employee Assistance Program
  • Training and development programs
  • Volunteering and donation matching program

Compensation

Base salary range (annual):

  • Zone A: $178,700 - $232,300
  • Zone B: $159,100 - $206,800
  • Zone C: $140,700 - $182,900

Beyond base salary, this role may be eligible for equity awards and participation in company incentive plans (such as annual discretionary bonuses or sales commissions). Final offer depends on geographic location and the candidate’s relevant experience and skills.

Additional details

  • Breadth that may help: experience with ad fraud, install/attribution fraud, or fraud in rewarded or incentivized advertising.
  • Detection methods that may help: unsupervised or semi-supervised detection such as autoencoders, isolation forests, or graph-based approaches to device and user linking.
  • MLOps that may help: migrating ML serving between platforms, or building MLOps tooling on Kubernetes.
  • Domain that may help: background in mobile ad tech, mobile games, or app monetization.

Eligibility and notice

  • Benefits eligibility, specific offerings, and coverage vary based on the country and employment status.
  • This position requires sufficient knowledge of English for professional verbal and written exchanges.
  • Under the San Francisco Fair Chance Ordinance, Unity will consider for employment qualified applicants with arrest and conviction records.
  • Headhunters and recruitment agencies may not submit resumes/CVs through this Web site or directly to managers.
  • Unity does not accept unsolicited headhunter and agency resumes.

Similar Jobs