Quesma - IT Press Tour #61 April 2025

Transcript

1. Quesma Jacek Migdal, CEO & Founder April 1, 2025 IT Press Tour, London
2. Jacek @ CEO, Founder of Quesma. Database gateway. Previously, 10 years at Sumo Logic. Joined early (~20) in SF Bay Area, opened Poland office in 2015. Reached $300M ARR, $1.7Bln+. NVIDIA CUDA and Meta intern.
3. Status Raised $2.5mln funding Nov 2023. 8 full-time employees. Shipped first product, landed enterprise customers. Demoing two next products April 2025.
4. Genesis
5. The year of exploration: My Journey to Founding Quesma In 2023 spend 9 months researching different business ideas by talking to customers. Sticked to my strengths. Start from big problem. See blog with same title.
6. Problem Structured data Modernize legacy: many horror migration stories Better insights: many decisions would be better if they were data-driven
7. Horror stories Organization Date TSB (UK Bank) April 2018 ~2 million customers could not use banking for a week due to failed migration, ~ £200 million in remediation costs and fine of £48.65 million (routers.com) reuters.com) Toyota August 2023 14 Toyota Plants closed for few days due to failed database maintenance (source) City of Denver July 2016 $4mln in extra licensing cost of Oracle due to audit (source)
8. Why? SQL is 50 years old • Paper 1970: A relational model of data for large shared data banks by Edgar F. Codd. • Information theory research laid the foundation for SQL. • Brought to the real world by Oracle with CIA funding.
9. Popularity of DB engines Source: db-engines.com Many new DB engines starting with slow traction No significant drop - customers are locked-in to old stacks. These are mostly FT500 logos
10. Why now? Exponential growth of data Push for efficiency AI is about data
11. Quesma Mission • The best insights through smarter querying of your data. • Allow customers to use best data tool for the job.
12. Why Quesma? • Greyfield data development • Database gateway ◦ Query proxy ◦ Intelligent transformations ◦ Blue/green deployments The Tottenham Court Road construction site (2009). carlbob, CC BY 2.0 via Wikimedia Commons
13. Quesma value proposition • Better data insights ◦ E.g. Reduce outage time • Lower cost ◦ E.g. 10 times better cost/perf
14. How Quesma works
15. How Quesma works • Transparent proxy between your application and databases: ◦ Interoperability ◦ A/B testing ◦ New query capabilities
16. Beachhead product #1: • Allow to use rich Elastic ecosystem with efficient data warehouse. ◦ Elasticsearch is designed for Google style searches ◦ 65%+ share of warehouse use case (Observability and Security). ◦ Expensive to maintain. 20 times slower than Clickhouse.
17. Demo
18. Use cases
19. Use case #1: G&L Systemhaus • Problem: Providing high quality streaming video, even during high-profile events. ◦ Spiky traffic, need real time visibility in huge volume. • Solution: Hydrolix with Quesma.
20. Use case #2: Major European telco • Successful proof of concept. • Problem: Monitoring and troubleshooting network quality. ◦ Very costly system. ◦ Could not keep up, lower productivity. • Solution: Quesma in air-gapped data center.
21. Use case #3: Major travel broker • Problem: Complex large scale system with many integrations. ◦ Costly system, underperforming. • Solution: Quesma with open-source ClickHouse
22. Competition and differentiators
23. Competition • $10 Bln / year data migration market (source) ◦ Mostly services (e.g. Accenture) • Database vendors provide tools to migrate to their platform or proxies to mitigate some shortcomings
24. Differentiation • Product offering. ◦ Gradual online approach to migration. ◦ Core expertise. • Independent of vendor ◦ Best tool for the job.
25. Roadmap
26. SQL with Pipe Syntax, Extensions nad AI • Likely OpenTelemetry Common Observability Query Language. • Based on Google Paper.
27. Quesma SQL with Pipe Syntax, Extensions and AI
28. Quesma charts Input: Ethnologue (2025) table from the Wikipedia "List of languages by total number of speakers" Prompt: "Plot top 10 languages, split by native and second-language speakers."
29. charts.quesma.com
30. Go to market • Source-available GitHub • Digital technical content • Outbound and conferences • Together with new database vendors
31. Pricing model • Source-available, free if your whole stack is open-source • Pricing aligned with other vendor ◦ $30 / TB / month of storage • Many enterprise deals
32. SQL with Pipe Syntax, Extensions nad AI • Likely OpenTelemetry Common Observability Query Language. • Based on Google Paper.
33. Soon April 2, 2025: Quesma at Notion Capital Cloud Challengers April 3, 2025: talk at KubeConEU “Observability Pipeline Query Languages: Present and the Future” April 22, 2025: AI Launchpad at Data Council, San Francisco Bay Area, California
34. Thank you! IT Press Tour Jacek Migdal <jacek@quesma.com>