About the Role
Join the Performance team at Vibe to own advertiser outcomes through prediction models, bidding, and delivery infrastructure. You will build core prediction architecture, manage a growing scale of data, and translate modeling efforts into measurable advertiser performance.
Responsibilities
- Own the prediction stack
- Improve the unified multi-task model for predicting visits, purchases, and outcomes
- Manage the full lifecycle of ideas from exploration to production including design, training, debugging, deployment, and monitoring
- Collaborate with Performance engineers on serving and bidding infrastructure
- Design new features from raw signal including household-level features
- Prepare the modeling stack for new, richer data sources
- Optimize for business impact and measurable uplift
- Iterate in production by shipping, measuring, and adjusting
Requirements
- Professional hands-on deep learning experience
- Strong Python proficiency
- Ability to diagnose model performance issues such as data leakage, bias, calibration, and distribution shift
- Ability to build algorithms from scratch
- Proven track record of framing work in business outcomes and KPIs
Skills
- Deep Learning
- Python
- PyTorch
- TensorFlow
- JAX
- Model debugging
- Algorithm development
- Data leakage analysis
- Model calibration
- Distribution shift analysis
Location
- Paris
- Hybrid (2-3 days in office)
Work Type
- Full-time
Experience Level
- Senior
Benefits
- Relocation support
About the Company
- Vibe.co is a streaming TV advertising platform that enables brands to reach audiences through hyper-targeted segmentation and AI-powered insights.
- The company is a fast-growing CTV platform that reached a $100 million revenue run rate in under two years.
- Backed by investors including Hedosophia, Elaia, Singular, QuantumLight, and Illusian.
