Serverless data ingestion at petabyte-scale with Amazon S3 presigned URLs
During the Data Science Summit we’ll show how we built a secure, serverless data ingestion service at petabyte scale. Powered by Amazon API Gateway, AWS Lambda, and Amazon S3 pre-signed URLs, our solution handles transfers of any size with enterprise-grade security, privacy, and full control over upload status.
Agenda
- Problem Statement & Requirements
- Evaluating Data Ingestion Solutions
- Amazon S3 Presigned URLs
Key topics
- Multi-tenant petabyte-scale data ingestion — enterprise security requirements, terabyte daily uploads, and full file processing status control;
- Architectural approaches — direct S3 access, Transfer Family, and API Gateway solutions with their trade-offs and limitations;
- Presigned URL strategy — PUT vs POST methods, single-part vs multi-part uploads, and size constraints;
- Serverless implementation — API Gateway with Lambda, security controls, and checksum validation for enterprise-grade data ingestion.
Speakers

Rafał Mituła AWS Community Hero
Head of DataHead of Data at Chaos Gears, wearing an Architect’s hat whenever needed. Recognized as an AWS Hero by AWS, Rafał actively co-organizes the AWS User Group Warsaw meetups and the AWS Community Day Poland conference.

Kamil Bartosik
Cloud Data EngineerHe focuses on building scalable, cloud-native solutions and enjoys working with both the infrastructure and data sides of projects. With experience in both cloud engineering and software development, he often contributes to architecture design and helps clients apply best practices in real-world projects.
Notes
- The event will take place at PGE National Stadium in Warsaw (Księcia Józefa Poniatowskiego 1).
- Tickets are limited and require registration.
- The Data Science Summit is the largest data conference in the CEE region, attracting over 2,500 participants. With a lineup of 200+ talks, the agenda covered the latest trends, cutting-edge solutions and real-world production experiences in fields like machine learning, data mining, big data, data management and quantum computing. Lectures were conducted in Polish and English.
- Any questions? Get in touch.