Dev Propulsion Labs

In this episode of Dev Propulsion Labs, Supabase CEO Paul Copplestone reveals why hiring ex-founders with beaten-down egos builds better products, how internal meme workshops became part of their culture, and why vibe coding isn't a bubble that will burst. He shares the accidental origin of Launch Weeks, explains why Supabase is building for a 30-year timeline, and breaks down how they scaled to 5 million developers across 40 countries with barely any meetings.Victoria Melnikova is a business and go-to-market expert for developer tools and AI, and host of Dev Propulsion Labs. She works with 40+ early-stage startups a year as Head of New Business at Evil Martians. https://x.com/vmelnikova_enEvil Martians is a design and engineering consultancy for developer tools, AI, and cybersecurity startups. https://evilmartians.com/Recorded at Chroma. Vector, full-text, regex, and metadata search. Develop locally and scale to petabytes in the cloud backed by object storage. Serverless search and retrieval that is fast, cheap, and reliable. https://www.trychroma.com

Sam Lambert, CEO of PlanetScale and expert in distributed databases, shares insights on surviving major AWS outages with extreme fault tolerance. He discusses the explosive growth of Vitess, emphasizing that operational excellence is more vital than flashy marketing. Sam also highlights the importance of community and events, along with how successful DevTools founders leverage product intuition and branding. Plus, he reveals the real costs of building reliable databases and advises against premature scaling while celebrating PlanetScale's rapid evolution.

Zeno Rocha, founder and CEO of Resend, discusses his journey from Brazil to building an $18M email API company. He shares insights on the importance of brand in attracting customers and talent, advocating for a zero-ego hiring approach. Zeno emphasizes cutting scope to deliver high-quality products quickly and the value of seeking rejection to accelerate sales. He also highlights the strategic role of open source and the need for a proactive customer focus in a competitive market.

Jeff Huber, Co-founder and CEO of Chroma, shares insights from his decade in applied machine learning. He discusses how small, opinionated teams with low egos create the best developer tools, emphasizing that consensus stifles innovation. Jeff critiques the RAG approach, introduces the concept of context engineering, and outlines Chroma’s strategy of keeping their core open source while monetizing complementary services. He also highlights the importance of design, company culture, and the mission to democratize AI-powered services.

Sarah Wooders, CTO and co-founder of Letta AI, dives into the intriguing world of stateful AI agents. She compares current LLMs to forgetting characters in 'Memento,' emphasizing their lack of memory. Discussing the AI landscape, she suggests 2025 mirrors the early internet's potential. Wooders critiques the distinction between true agents and mere marketing hype, while also stressing the necessity for better standardization in AI protocols. Open-source tools, she argues, are vital for fostering genuine advancements in AI capabilities.

In a compelling discussion, Adam Frankl, a renowned author and partner at Alchemist Accelerator, shares why 2025 is an ideal year to launch a developer tool startup amid the chaos of AI-driven changes. He highlights that founders should focus on the problem rather than the product, emphasizing the importance of becoming an authority on specific challenges. Adam also advises utilizing social media strategically to build authority and engage with the developer community, while pointing out that enterprise clients are eager to invest in AI solutions.

Jason Bosco, CEO and co-founder of TypeSense, shares how he and his co-founder built a profitable search engine serving 10 billion searches monthly without taking VC funding. From Dollar Shave Club VP of Engineering to bootstrapped founder, Jason reveals the unconventional path to building sustainable developer tools.Key insights from this episode:"If it is hard for you as a founder to convince someone to pay you, it's never gonna get easier from there." Find what people are willing to pay for early - don't build first and monetize later."We're opinionated and we want search to work out of the box right from the get-go." TypeSense chose simplicity over configurability, targeting 80% of use cases with zero-config search versus Elasticsearch's thousands of parameters."We don't want the gamble on TypeSense the company to end up affecting TypeSense the product." Jason explains why they chose profitability over VC funding to build a multi-generational product without the pressure of 10x returns."Doing dev tools in closed source is like playing it on hard mode." Open source creates better feedback loops with developers, leading to faster product iteration and stronger community adoption.Links:- TypeSense: https://typesense.org/- TypeSense Cloud: https://cloud.typesense.org/- Jason Bosco on X: https://x.com/jasonbosco- Evil Martians on X: https://x.com/evilmartians- Victoria Melnikova on X: https://x.com/vmelnikova_enEvil Martians is the go-to agency for early-stage developer tools startups: https://evilmartians.com/devtools

Anna Veronika Dorogush, founder and CEO of Recraft, reveals how she built one of the world's leading AI image generation platforms by solving real professional design problems instead of chasing AI hype. Some key insights:"My whole back-end team is medalists and finalists of World Championship in programming." Strong people attract strong people, creating a talent density that enables a small team to compete with giants."We are just focused on producing the best models in image generation space for designers, for professional use cases." While others built general AI image generators, Recraft targeted designers' specific needs: brand consistency, style control, and professional workflows. "That's the major differentiator between ourselves and other AI native tools is we are building our technology from scratch in-house. And that allows us to solve for professional tasks." Training proprietary models in-house allows solving for your users' exact problems (controlling styles, brand colors, fonts)."At the first stage, think investors mostly are evaluating founders and founding teams. After that, investors are evaluating product market fit and retention and later, monetization starts to be very important. We've raised three rounds so far and on every one of those rounds, different things were considered very important." Links:Recraft: https://www.recraft.ai/Anna Veronika on X: https://x.com/avwritingEvil Martians on X: https://x.com/evilmartiansVictoria Melnikova on X: https://x.com/vmelnikova_enEvil Martians is the go-to agency for early-stage developer tools startups: https://evilmartians.com/devtools

Evil Martians is the go-to agency for early-stage developer tools startups: https://evilmartians.com/devtoolsLinks:- tambo ai: https://tambo.co/- Michael Magán on X: https://x.com/mrmagan_- Evil Martians on X: https://x.com/evilmartians- Victoria Melnikova on X: https://x.com/vmelnikova_en

José Valim, the creator of Elixir and founder of Dashbit, shares insights from his journey in developing a beloved programming language. He discusses how prioritizing personal curiosity over market trends led to genuine community engagement. Valim emphasizes decentralization in the Elixir ecosystem, enabling diverse innovation. He also highlights the importance of transparent marketing through clear trade-offs. Currently, he's focused on Tidewave, aiming to create intuitive AI tools that enhance web development experiences.