Data Engineer – Mobile Monetization – Big Data, GenAI & Mobile AdTech

Responsibilities

Design and build highly scalable data platforms to process billions of mobile advertising events with low latency and high availability. Collaborate with cross-functional teams to implement real-time streaming pipelines, batch processing jobs, and GenAI-powered analytics solutions.

Requirements

Requires 1–5+ years of professional experience in Java backend development and data engineering with strong knowledge of distributed systems. Candidates must have hands-on experience with Big Data technologies like Spark, Kafka, and Hadoop, along with proficiency in Go or Rust.

Skills

• Java
• Data engineering
• Spark
• Kafka
• Hadoop
• Go
• Rust
• SQL
• Distributed systems
• Real-time analytics
• Generative AI
• Mobile AdTech
• ETL/ELT
• Data modeling
• Cloud infrastructure
• API development

Benefits

• Paternity leave
• Maternity leave
• Healthcare insurance
• Broadband reimbursement
• Healthy snacks and drinks
• Catered lunches

About this role

About the Role
PubMatic is seeking for a passionate Data Engineer to join our Mobile Advertising Platform engineering team. In this role, you will build highly scalable data platforms powering Mobile Programmatic Advertising, RTB bidding systems, attribution, campaign optimization, fraud detection, and real-time analytics.
You will design and implement distributed systems capable of processing billions of mobile advertising events every day including ad requests, impressions, clicks, installs, conversions, in-app events, and auction logs. You'll build both real-time event processing and large-scale batch processing systems on PubMatic's in-house data centers, leveraging modern Big Data technologies (Spark, Kafka, Hadoop), high-performance services in Go and Rust, and Generative AI to deliver intelligent analytics and automation solutions.
You'll collaborate with Product Managers, ML Engineers, Data Scientists, and Backend Engineers to build next-generation Mobile AdTech products.
What You'll Do:
Design and build highly scalable data platforms capable of processing billions of mobile advertising events with low latency and high availability.
Design and build ultra-low-latency, high-throughput data pipelines processing the full mobile event funnel — ad request → impression → click → install → in-app conversion event — including OpenRTB bid request/response logging, auction analytics, and win-rate/bid-landscape reporting across multiple SSPs and ad exchanges, using Kafka, Spark Streaming, and Hadoop on our in-house data centers.
Architect real-time campaign pacing and budget delivery systems, applying bid optimization techniques so spend controls and bid strategy adjustments react to live auction signals within seconds.
Architect and implement real-time streaming pipelines to ingest, process, and analyze billions of advertising events including impressions, clicks, installs, conversions, and in-app events.
Build real-time event processing systems with event-time semantics, windowed aggregations, watermarking, late/out-of-order event handling, deduplication, and exactly-once or idempotent delivery guarantees across Kafka and Spark Structured Streaming.
Design and optimize scalable ETL/ELT pipelines and data models supporting Mobile AdTech use cases such as campaign reporting, attribution, audience segmentation, fraud detection, and revenue analytics.
Develop and tune large-scale batch processing jobs on Spark and Hadoop (HDFS/YARN) — hourly and daily aggregations, backfills, reprocessing, and data reconciliation between real-time and batch layers — with strong workflow orchestration, SLAs, and data-quality checks.
Design large-scale Invalid Traffic (IVT) and ad fraud detection systems — click spam, install hijacking, SDK spoofing, device farms — using rules-based and ML-driven detection across billions of daily events.
Develop high-performance backend services and data processors (Java, Go, Rust, REST APIs, JDBC) supporting DSP bidding logic, campaign management tools, and reporting APIs.
Operate and scale data infrastructure in PubMatic's in-house data centers — Kafka, Spark, and Hadoop clusters — including capacity planning, performance tuning, storage and compute efficiency, and cost optimization on bare-metal hardware.
Design and develop GenAI-powered agents for analytics, operations, and data enrichment using frameworks such as LangChain, LlamaIndex, LangGraph, or custom orchestration frameworks.
Design systems that are highly available, fault tolerant, and optimized for low latency and high throughput.
Collaborate closely with cross-functional teams to improve platform scalability, reliability, performance, and customer experience.
Work with Product Managers to define and implement new analytics capabilities and platform features.
Participate in Agile/Scrum ceremonies including sprint planning, backlog grooming, estimation, retrospectives, and release planning.
Perform architecture discussions, design reviews, and peer code reviews while promoting software engineering best practices.
We'd Love for You to Have:
1–5+ years of professional experience in Java backend development and data engineering.
Strong computer science fundamentals including data structures, algorithms, distributed systems, and software design principles.
Hands-on experience developing scalable backend services using Java, REST APIs, JDBC, and relational databases.
Strong experience building distributed real-time and batch data pipelines using Spark, Kafka, Hadoop, and SQL on on-premise / in-house data center infrastructure.
Experience designing and optimizing large-scale analytical data platforms for Mobile AdTech using Java, Go, Rust, Spark, Kafka, and SQL. Strong understanding of real-time event processing, distributed data systems, and data warehouse architectures. Familiarity with distributed query engines, modern data lake technologies, and high-performance analytical databases (e.g., Trino, Presto, ClickHouse, Apache Iceberg, Delta Lake, or Apache Hudi) is an added advantage.
Experience processing high-volume streaming data using Spark Streaming, Kafka Streams, or similar technologies.
Programming proficiency in Go and/or Rust for building high-throughput, low-latency data services, event collectors, and stream processors, in addition to Java.
Solid understanding of real-time vs. batch processing trade-offs (Lambda/Kappa architectures), event-time processing, partitioning strategies, schema evolution (Avro/Protobuf/Parquet), and data correctness at scale.
Good understanding of data warehousing, dimensional modeling, ETL/ELT design, and distributed computing concepts.
Excellent verbal and written communication skills.
Strong interpersonal skills with the ability to collaborate effectively across engineering, product, and customer-facing teams.
Bonus Qualifications
Mobile AdTech Experience
Experience working with Mobile AdTech platforms or digital advertising ecosystems is highly p...

About PubMatic

Technology, Information and Internet · 501-1,000 employees · headquartered in Redwood City, CA · founded 2006