PubMatic is looking to solve the problem of optimizing campaign performance through high-throughput, real-time systems in the performance advertising platforms
Requirements
- Strong programming skills in Java or Scala (Python a plus)
- Hands-on experience with real-time streaming frameworks (Kafka Streams, Flink, Spark Streaming) and distributed systems
- Familiarity with programmatic advertising protocols (OpenRTB, VAST) and campaign delivery workflows
- Proficiency in SQL and experience with large-scale analytics platforms (Snowflake, BigQuery, or similar)
- Practical experience in low-latency system design and performance tuning
- Exposure to attribution modeling, incrementality measurement, and performance KPIs (CPA, ROAS, CTR, CVR)
- Understanding of cloud platforms (AWS, GCP, Azure) and container orchestration (Docker, Kubernetes)
Responsibilities
- Design, build, and optimize low-latency, fault-tolerant real-time systems for ad serving, bidding, and performance optimization
- Develop high-throughput data pipelines for click, conversion, and attribution events using Kafka, Spark, or Flink
- Implement real-time bidding algorithms, pacing, budget management, and supply path optimizations
- Work with deal curation and audience targeting components, integrating with forecasting and attribution services
- Develop backend services and APIs for activation, reporting, and optimization workflows
- Collaborate with ML engineers to productionize models (e.g., bid shading, contextual bandits, reach forecasting)
- Integrate with major programmatic protocols (OpenRTB, VAST, CTV-specific standards)
Other
- Possess 4+ years of experience in software engineering with a focus on ad tech / performance advertising systems
- Proven ability to own projects end-to-end and deliver high-quality, scalable solutions
- Experience working closely with product managers, data scientists, and analytics teams in ad tech
- Strong foundation in data structures, algorithms, and distributed architecture
- Return to Office: PubMatic employees throughout the globe have returned to our offices via a hybrid work schedule
