3 BILLION searches per month without VC funding - Jason Bosco from Typesense
Typesense founder Jason Bosco explains how a nights-and-weekends search project became a profitable company serving three billion searches a month—and why “revenue funded” is a permanent model, not a waiting room for venture capital.
Jason Bosco is the founder of Typesense, an open-source, batteries-included search API positioned as an alternative to Algolia.
He lays out the capital, patience and incentives behind a revenue-funded DevTool: eight years of product maturation, demand before the hosted launch, organic integrations driven by users and a company whose only paying constituency is its customers.
The Q&A below is experimental — extracted from the episode transcript with AI. If you spot a mistake, let us know at jack@scalingdevtools.com.
What does it mean to build Typesense without venture capital?
Every company needs capital; the choice is where it comes from and what expectations follow. We invested nights and weekends from 2015, used revenue from other side projects during the transition and kept our jobs until self-hosting users said they would pay for a hosted product. When Typesense Cloud launched in 2020, customers arrived in the first week. Our time and prior products supplied non-dilutive capital before revenue took over.
Why did the product take five years before monetization?
Typesense began as R&D into why search was so complicated. From 2015 to 2018 we tested algorithms, benchmarks and an in-memory architecture that could simplify Elasticsearch’s model. We then open-sourced it, launched on Hacker News and Product Hunt, and spent another two years learning from users. That gradual path meant the product and its demand existed before we depended on it for our livelihoods.
What baggage can venture capital add to a DevTool?
A fund eventually needs growth and an exit, which can push a company toward enterprise-only features, opaque pricing, higher renewals and the largest accounts. The company is effectively selling two products—to users and stock to investors—and the second customer group can influence the first. Those incentives are understandable, but they can turn the tool away from the smaller developers who helped it mature.
How does being revenue funded change daily decisions?
When I wake up, I only need to ask what customers are telling us and which use cases we should support next. There is no separate stock product whose buyers require a different trajectory. That alignment creates contentment and gives an infrastructure product time to become dependable over a decade or more, rather than forcing every product choice through the valuation growth expected by another constituency.
How did Typesense earn a place in Firebase and Laravel ecosystems?
Firebase users adopted Typesense, improved it through feedback and publicly asked why it was missing from the documentation. The push came from a critical mass of users, not an outreach campaign. Laravel followed a similar path: we maintained a third-party integration, improved it with real usage and eventually had enough scale and support capacity to contribute it upstream. Trust accumulated over years before the official integrations became possible.
Do infrastructure founders still need to move fast?
Speed is relative to the problem. Search has existed for decades and will remain necessary wherever data must be retrieved, so no single quarter closes the opportunity. We must keep pace with customer expectations, hardware and machine learning, but a calmer release cadence can reduce integration fatigue. Time-sensitive or capital-intensive markets may need VC; evergreen infrastructure can often benefit from a longer runway.
Why call the company “revenue funded” instead of bootstrapped?
Bootstrapped is often heard as an early stage before a company raises money, even by very young startups evaluating an established, profitable vendor. Revenue funded describes a durable state: customers finance the company and it remains profitable. We were bootstrapped while supplying the initial capital ourselves; now we are revenue funded. The language makes clear that VC is not the assumed next step.
Where is venture capital still the right tool?
Capital-intensive work such as training models or acquiring GPUs may be extremely difficult to fund organically, and an opportunity with a short window may reward rapid scaling. I am glad venture capital exists for those experiments. My argument is narrower: software DevTools often need time to mature, and founders should consider whether VC incentives distort that natural progression before treating it as the universal default.
What is the one takeaway for founders?
Treat venture capital as one option among many. Consulting, an agency, other software products or paid information can fund the period before the main product supports itself. The right source depends on the business, but founders should creatively compare all of them rather than assume that a company with large impact can be built through only one financing path.