Skip to content
YC Root AccessYC Root Access

Supabase: Cash Does Not Equal Success

In this fireside from Startup School Paris, Supabase CEO Paul Copplestone reflects on creating one of the fastest growing dev tools in the AI era. Apply to Y Combinator: https://www.ycombinator.com/apply Work at a startup: https://www.ycombinator.com/jobs Chapters: 00:00 — Paul's Origin Story 01:56 — Building Software He Needed 15 Years Ago 03:11 — Why Bet on 30-Year-Old Postgres? 04:17 — The Open Source Decision 07:17 — Do Cloud Giants Threaten You? 08:33 — The "Open Source Firebase" Pivot 09:59 — Missing Features, Big Ambitions 10:26 — Winning Developers Over Firebase 12:18 — Launch Waves & YC Batch Adoption 14:27 — How Agents Changed the Metric 16:50 — Building the First Docs Chatbot 21:18 — Claude Code Changes Everything 22:55 — 60%+ of Databases Launched by Agents 27:03 — Scaling from 220 to 360 People 31:17 — AI Inside the Company 34:19 — True Gains Need No Human in the Loop 38:43 — $500M Round, Staying Grounded 40:47 — Advice for DevTool Founders Today

Paul Copplestoneguest
Aug 21, 202641mWatch on YouTube ↗

At a glance

WHAT IT’S REALLY ABOUT

Supabase’s agent-driven growth proves cash doesn’t equal real success

  1. Paul Copplestone explains why Supabase bet on Postgres: an increasingly popular, community-owned database whose ecosystem flywheel makes it a durable long-term foundation.
  2. Supabase committed to truly permissive open source (MIT/Apache/Postgres) and monetizes by running the hard operational parts of Postgres rather than withholding features behind an open-core model.
  3. The company’s breakout came from crisp positioning as an “open source Firebase alternative,” then winning developers on predictable pricing, relational data, and open-source credibility.
  4. Supabase’s growth accelerated across three AI waves—pgvector/RAG, vibe-coding platforms launching massive numbers of small databases, and agentic coding (Claude Code) driving faster activation and conversion.
  5. Agent-driven usage is reshaping Supabase’s DX priorities toward CLI/MCP-first tooling, code-defined workflows, and tighter dashboard-to-git synchronization while the company scales from ~220 to ~360 people.

IDEAS WORTH REMEMBERING

5 ideas

Postgres was the “old tech” bet precisely because it’s community-owned and compounding.

Copplestone framed Postgres’ durability as an ownership advantage: because no single company controls it, contributions and adoption compound over time. That ecosystem “flywheel” makes Postgres a stable foundation for building a long-lived database business.

Supabase monetizes operational complexity, not proprietary features.

Supabase chose permissive licenses (MIT/Apache/Postgres) and avoided open-core gating, betting that the managed-service difficulty (backups, failover, ops) is the natural monetization. Their differentiation is building an ecosystem around Postgres and contributing upstream rather than inventing a proprietary “Postgres-like” database.

A clear category frame (“open source Firebase”) unlocked distribution more than raw features did.

Early traction was limited under the “real-time Postgres” positioning; changing the tagline to “open source Firebase alternative” made the value prop instantly legible and triggered a top-of-Hacker-News breakout. Even without feature parity, the aspirational comparison clarified who the product was for and why it mattered.

Supabase’s wedge against Firebase was trust + relational data + predictable economics.

They won developers by addressing common Firebase pain points: pricing/bill shock, NoSQL limitations, and Google lock-in sentiment—while offering relational Postgres, predictable pricing, and true open source. Over time this credibility shift is reflected in adoption (e.g., a claimed 60%+ of a recent YC batch using Supabase).

AI didn’t just add new users—it changed the quality and lifecycle speed of Supabase workloads.

Copplestone described three AI-driven growth waves: (1) pgvector/RAG adoption, catalyzed by an outside contributor who added pgvector and helped build an early “chat with docs” demo; (2) the vibe-coding platform wave (Lovable/Bolt) creating massive database-creation bursts; (3) Claude Code/agentic coding driving higher-intent builds where activation, conversion, and revenue curves all rose together.

WORDS WORTH SAVING

5 quotes

big problems, big, hairy problems present good opportunities for a startup founder.

Paul Copplestone

Postgres belongs to no one.

Paul Copplestone

on one day, we changed the tagline from real-time Postgres to open source Firebase alternative.

Paul Copplestone

over 60%, uh, of databases are launched by an agent.

Paul Copplestone

staying grounded is largely a factor of just remembering that the cash actually doesn't mean that you're successful.

Paul Copplestone

Founder origin story and prior startupsBetting on Postgres adoption trendsPermissive open-source licensing vs open corePositioning pivot to “open source Firebase”Developer experience and time-to-value metricAI waves: pgvector/RAG, vibe coding, Claude CodeAgent-launched databases and CLI/MCP-first tooling','Scaling headcount, reliability, and org processes','AI-enabled internal operations (Slack/Notion, Hex)','Fundraising, operating capital, and staying grounded

High quality AI-generated summary created from speaker-labeled transcript.

Get more out of YouTube videos.

High quality summaries for YouTube videos. Accurate transcripts to search & find moments. Powered by ChatGPT & Claude AI.