Rootconf Topical Edition on Databases

A practitioner-focused edition on modern database systems, bringing production engineers and database researchers to examine how modern data systems are actually designed, scaled, tested, and operated in production, from transactional databases and control planes to query engines, distributed correctness, and AI-era workloads.

Submissions and participation are already in from teams at Microsoft, Antithesis, Nutanix and Feldera, with a pipeline of strong submissions from AWS, Yugabyte, CockroachDB, InMobi, and Teradata.

Keynote announcement

Database systems: a decade of disruption and innovation by Karthik Ramachandra (Azure SQL at Microsoft)

Over the last decade, databases have evolved from tightly controlled on-prem systems into elastic, cloud-native services operating at planetary scale.

This keynote traces the systems innovations that enabled this transformation - from large-scale query processing and cloud-native architectures to multi-modal workloads, governance, and distributed data management.

The keynote concludes by examining the next wave of challenges and opportunities for databases in the era of AI.

Key takeaways

• How database systems evolved over the last decade
• Deep systems innovations that reshaped industry and academia
• Open problems and opportunities emerging in the AI era

Newly selected talks

Rolling Your Own Database (Safely!)

Property-based testing at scale - Shomik Ghose (Antithesis)

How three engineers built and validated a production OLAP database using property-based testing, deterministic simulation, and fuzzing, uncovering correctness bugs that conventional testing would never catch.

Topics include:

• Property-based testing for databases
• Deterministic simulation
• Correctness verification
• Production-grade OLAP systems

Beyond Polling

Event-driven multi-cluster database control planes - Vaibhaw Pandey and Marko Nikolic (Nutanix)

A deep dive into designing scalable, event-driven state synchronization for PostgreSQL control planes running across multiple Kubernetes clusters.

The talk covers:

• Kubernetes Informers at scale
• Reconnection semantics and drift recovery
• Leader election patterns
• Multi-cluster state synchronization

Previously announced highlights

Rethinking incremental computation - Mihai Budiu (Feldera)

A deep dive into Incremental View Maintenance (IVM) and how Feldera applies it to deliver dramatic improvements in query latency and compute efficiency.

Why databases were not designed for agents - Arpit Bhayani (Razorpay)

When LLM agents start talking to your database, assumptions around queries, schemas, and reliability slowly apart.

Fast on paper, slow in reality - Sarthak Makhija (Caizin)

Lessons from building a distributed key-value store and the production bottlenecks that emerge beyond “correct” designs.

Editorial team

Curated by practitioners building and operating large-scale data systems.

Editor

Sandeep Joshi (Uber)

Reviewers

• Anirudh Rowjee (Couchbase)
• Varuni HK (Couchbase)
• Kumar Abhijeet (Index Exchange)
• Ranganadh Thatha (Mico)
• Rajkumar Iyer (Microsoft)
• Siddhartha Reddy (Udaan)
• Raj Suvariya (Stripe)
• Rolland Santimano (Rubrik)
• Rohan Reddy Alleti (Sahaj)
• Mohit Gurnani (Nutanix)
• Varun Mishra (Flipkart)
• Talina Shrotiya (Couchbase)
• Abdul Samad (Flipkart)

Agenda

Friday, 12 June

Hands-on workshops, followed by mixers and community gatherings across Bengaluru.

Saturday, 13 June

Conference at TERI Auditorium, Indiranagar.

Attend

🎟 Conference tickets and annual memberships are now open.

Sponsorship inquiries

sales@hasgeek.com

Contact information

📞 Call or WhatsApp Rootconf: +91 76763 32020
📧 Email: info@hasgeek.com