/
© 2026 RiffOn. All rights reserved.
  1. A Product Market Fit Show | Startup Podcast for Founders
  2. He built a new database in his bedroom—now he powers Cursor, Notion and Anthropic. | Simon Eskildsen, Founder of turbopuffer
He built a new database in his bedroom—now he powers Cursor, Notion and Anthropic. | Simon Eskildsen, Founder of turbopuffer

He built a new database in his bedroom—now he powers Cursor, Notion and Anthropic. | Simon Eskildsen, Founder of turbopuffer

A Product Market Fit Show | Startup Podcast for Founders · Oct 30, 2025

How Simon built TurboPuffer, a 100x cheaper vector DB, landed Notion & Cursor, and raised capital with a clear, principled approach.

First-Principles "Napkin Math" Uncovers Opportunities Incumbents Miss

The founder used a "Napkin Math" approach, analyzing fundamental computing metrics (disk speed, memory cost). This revealed a viable architecture using cheap S3 storage that incumbents overlooked, creating a 100x cost advantage for his database.

He built a new database in his bedroom—now he powers Cursor, Notion and Anthropic. | Simon Eskildsen, Founder of turbopuffer thumbnail

He built a new database in his bedroom—now he powers Cursor, Notion and Anthropic. | Simon Eskildsen, Founder of turbopuffer

A Product Market Fit Show | Startup Podcast for Founders·4 months ago

High-Stakes Enterprise Deals Are Won by "Willing Them Into Existence"

To land Notion, co-founder Justin worked with extreme intensity, finding 300ms of latency in three hours and building a requested feature in 24 hours after promising it on the spot. This level of obsessive commitment is required to win transformative customers.

He built a new database in his bedroom—now he powers Cursor, Notion and Anthropic. | Simon Eskildsen, Founder of turbopuffer thumbnail

He built a new database in his bedroom—now he powers Cursor, Notion and Anthropic. | Simon Eskildsen, Founder of turbopuffer

A Product Market Fit Show | Startup Podcast for Founders·4 months ago

Win Your First Major Customer By Flying to Their Office and Solving Their Problems

To land an unresponsive prospect, the founder flew to their office. He arrived as they were fighting a database fire and immediately helped them fix it. This impromptu help session proved his expertise and built immense trust that led them to become a customer.

He built a new database in his bedroom—now he powers Cursor, Notion and Anthropic. | Simon Eskildsen, Founder of turbopuffer thumbnail

He built a new database in his bedroom—now he powers Cursor, Notion and Anthropic. | Simon Eskildsen, Founder of turbopuffer

A Product Market Fit Show | Startup Podcast for Founders·4 months ago

True Pricing Disruption Is When Customers Worry You'll Go Out of Business

After TurboPuffer quoted a price, Notion's team asked if they would lose money on the deal. This concern wasn't a negotiation tactic but a genuine fear that their new critical vendor was unsustainable. This is a powerful signal of true market disruption.

He built a new database in his bedroom—now he powers Cursor, Notion and Anthropic. | Simon Eskildsen, Founder of turbopuffer thumbnail

He built a new database in his bedroom—now he powers Cursor, Notion and Anthropic. | Simon Eskildsen, Founder of turbopuffer

A Product Market Fit Show | Startup Podcast for Founders·4 months ago

Startups Should Only Fundraise for One of Six Legitimate Reasons

The founder classifies fundraising into six buckets: finding PMF, funding growth, employee liquidity, trust/publicity, strategic partnerships, or ego. This framework helps founders avoid raising capital for momentum's sake, which often adds unnecessary risk and dilution.

He built a new database in his bedroom—now he powers Cursor, Notion and Anthropic. | Simon Eskildsen, Founder of turbopuffer thumbnail

He built a new database in his bedroom—now he powers Cursor, Notion and Anthropic. | Simon Eskildsen, Founder of turbopuffer

A Product Market Fit Show | Startup Podcast for Founders·4 months ago

A 100x Cost Reduction Is Possible by Accepting One Non-Critical Performance Trade-Off

TurboPuffer achieved its massive cost savings by building on slow S3 storage. While this increased write latency by 1000x—unacceptable for transactional systems—it was a perfectly acceptable trade-off for search and AI workloads, which prioritize fast reads over fast writes.

He built a new database in his bedroom—now he powers Cursor, Notion and Anthropic. | Simon Eskildsen, Founder of turbopuffer thumbnail

He built a new database in his bedroom—now he powers Cursor, Notion and Anthropic. | Simon Eskildsen, Founder of turbopuffer

A Product Market Fit Show | Startup Podcast for Founders·4 months ago

Generational Database Companies Require a New Workload and New Storage Architecture

To build a multi-billion dollar database company, you need two things: a new, widespread workload (like AI needing data) and a fundamentally new storage architecture that incumbents can't easily adopt. This framework helps identify truly disruptive infrastructure opportunities.

He built a new database in his bedroom—now he powers Cursor, Notion and Anthropic. | Simon Eskildsen, Founder of turbopuffer thumbnail

He built a new database in his bedroom—now he powers Cursor, Notion and Anthropic. | Simon Eskildsen, Founder of turbopuffer

A Product Market Fit Show | Startup Podcast for Founders·4 months ago

Founders Must Trust Intuition Over Even the Most Articulate VC Advice

Founders with deep market fit must trust their unique intuition over persuasive, but generic, VC advice. Following the standard playbook leads to cookie-cutter companies, while leaning into the 'weird' things that make your business different is what creates a unique, defensible moat.

He built a new database in his bedroom—now he powers Cursor, Notion and Anthropic. | Simon Eskildsen, Founder of turbopuffer thumbnail

He built a new database in his bedroom—now he powers Cursor, Notion and Anthropic. | Simon Eskildsen, Founder of turbopuffer

A Product Market Fit Show | Startup Podcast for Founders·4 months ago