About the Role
As a Data Scientist on Taboola’s R&D Team in Downtown Los Angeles, utilizing some of the latest technology and tools, you will work on the recommendation and ranking algorithms that drive Realize — Taboola’s performance advertising platform serving 600M+ daily active users — and building agentic AI systems alongside it. The models you build directly influence campaign performance and revenue outcomes for advertisers worldwide.
How you’ll make an impact:
- Design and build predictive customer behavior models for targeting and personalization
- Implement the applicable Machine Learning or statistics based algorithm for prediction and optimization and deliver the trained model to production
- Take end-to-end ownership: design, build, ship, measure, and maintain models and the services around them—rapidly prototyping and delivering complete projects/products using AI-assisted “vibe coding” workflows.
- Act as a Harness Engineer: build the evaluation harnesses, automated testing frameworks, benchmarks, and guardrails that let the team experiment, measure model quality, and ship AI/ML systems safely and repeatedly, following a TDD process.
- Present findings to the product team and technical team leads in a clear and actionable way.
- Build and maintain code to populate HDFS, Hadoop with log from Kafka or data loaded from SQL production systems
- Design, build and support algorithms of data transformation, conversion, computation on Hadoop, Spark and other distributed Big Data Systems
To thrive in this role, you’ll need:
- Master’s degree or Ph.D, ideally in Computer Science, Electrical Engineering, Physics, Computational Economics, or similar
- 5+ years of hands-on experience in Python and Java
- Proficiency with ML frameworks (e.g., PyTorch, Tensorflow, Spark MLlib)
- SQL proficiency for data exploration and analysis
- System-architecture mindset — ability to design scalable, reliable, high-performance systems and reason about how a model fits into the larger data and serving infrastructure.
- Experience taking end-to-end ownership of ML systems: design, build, ship, measure, and maintain — using AI-assisted “vibe coding” tools (e.g., Cursor, Claude, GitHub Copilot).
- Experience building automated evaluation harnesses and test frameworks for ML systems — defining benchmarks, baselines, and guardrails that let a team ship confidently and repeatedly
- Ability to communicate technical findings clearly, and to collaborate effectively across teams (including early morning or evening syncs with our Tel Aviv team)
Bonus points if you have:
- Experience with recommendation systems, ranking algorithms, or personalization at scale
- Familiarity with real-time ML and large-scale data mining