Milind Srivastava
RMCIC 2224B
4720 Forbes Ave
Pittsburgh, PA 15213
I’m a PhD student at Carnegie Mellon University, and am extremely fortunate to be advised by Vyas Sekar. I work on building systems that leverage sketch-based approximation to reduce the cost and improve the interactivity of data analytics and observability workloads. I am a part of Laude Slingshots.
I am one of the leads and core architects of ProjectASAP, a grand vision project to democratize the benefits of approximate summaries for big data analytics workloads. We released ASAPQuery, a drop-in acceleration and cost reduction system for SQL and PromQL; and FlinkSketch, a library of sketching algorithms for the Flink streaming engine. Stay tuned for more open source releases from ProjectASAP :)
Previously, I worked on network telemetry and network security research. I did my undergrad in Computer Science at IIT Madras, where I worked on hardware security and side channel attacks. Outside work, I love cooking, biking, and hiking. I’m also an aviation nerd and would like to get a pilot’s license someday.
Recent Talks
- [Mar 2026] ASAPQuery: A Drop-in System to Accelerate Metrics Obserability by 100x using Sketch-based Approximation - Observability Day Europe 2026 (video coming soon)
- [Oct 2025] FlinkSketch: Democratizing the Benefits of Sketches for the Flink Community - Current 2025 [Video and Slides]
- [Oct 2024] Reducing Your AWS Analytics Cost by 10x Using Principled Approximation - Current 2024 [Video and Slides]
- [Aug 2024] Re-imagining Network Telemetry from an Approximation-First Perspective - SIGCOMM SRC 2024 (Awarded first prize) [Abstract]