AI & Agents

AI coding tools, agents, LLMs, and the future of AI in developer tools.

47:58Episode 160

How can you actually use AI in DevTools content? With Victor Coisne from Strapi

Victor, VP of Marketing at Strapi, walks us through how AI can be used in content creation—what tools work, what to watch out for, and how you can try some of these techniques yourself.

This episode is brought to you by WorkOS. If you're thinking about selling to enterprise customers, WorkOS can help you add enterprise features like Single Sign On and audit logs.

Links:

   •  Victor's X
   •  Victor's Linkedin
   •  Strapi
   •  GrowthX
   •  Kapa
   •  Octolens
   •  Semrush

39:33Episode 156

AI Tools for Enterprise - Chris and Matt from Ona

Gitpod has rebranded to Ona and shifted its focus to building AI tools for enterprise teams. This episode digs into why they made the leap, how they're standing out in a crowded AI space, and what it’s been like rethinking developer workflows from the ground up. We talk about dev environments, differentiating in the AI space, forward-deployed engineers and more.

This episode is brought to you by WorkOS. If you're thinking about selling to enterprise customers, WorkOS can help you add enterprise features like Single Sign On and audit logs.

Links:

   •  Ona
   •  Christian's X
   •  Matthew's X

44:22Episode 154

Karan Vaidya, founder of Composio: MCP use cases & Elon retweets

Karen from Composio shares how developers are using MCP to connect tools like Slack, Notion, and Gmail with AI agents, growing from nearly zero to 100,000 users in 6 months. They capitalized on key moments when new AI tools, such as Grok versions and Claude releases, came out, creating examples and demos that resonated strongly across social media and got them retweeted by Elon Musk. Hear how the team learns to use these tools better over time, helping each new release work smarter than the last.

This episode is brought to you by WorkOS. If you're thinking about selling to enterprise customers, WorkOS can help you add enterprise features like Single Sign On and audit logs.

Links:
   •  Composio
   •  Composio's X
   •  Karan's X
   •  Launch Video

36:13Episode 153

Studying Lee Robinson, Cursor's new VP of Developer experience

Lee Robinson helped Vercel grow to $200M+ in ARR and scaled the Next.js community to over 1.3 million active developers. I dive into his blog posts to uncover valuable insights and lessons about how he achieved this success, covering topics like docs, community building, developer education, marketing, and product development.

This episode is brought to you by WorkOS. If you're thinking about selling to enterprise customers, WorkOS can help you add enterprise features like Single Sign On and audit logs.

Links:
   •  Lee Robinson's blog
   •  Lee Robinson's X
   •  Peter Yang's interview
   •  swyx's interview
   •  Gonto on Scaling DevTools
   •  Developer Marketing Community

P.s. this is a new style of episode, let me know what you think. 
44:42Episode 150

Matt Palmer on Replit's speedrun to $100M ARR

Matt Palmer from Replit shares how the company scaled to $100M in ARR from ~$10M in under a year. We talk about the importance of video for teaching the non-linear process of working with AI, the challenge of rewriting documentation for a broader audience using the Diátaxis framework, and how they support a diverse community of users navigating this new AI-driven development landscape.

This episode is brought to you by WorkOS. If you're thinking about selling to enterprise customers, WorkOS can help you add enterprise features like Single Sign On and audit logs.

Links:
37:51Episode 149

Logan Kilpatrick from Google DeepMind: Building for 100m developers

Logan Kilpatrick shares how DeepMind's organizational changes helped their resurgance in AI. What needs to happen to reach 100m developers. And why the next six months are more exciting than ever.

This episode is brought to you by WorkOS. If you're thinking about selling to enterprise customers, WorkOS can help you add enterprise features like Single Sign On and audit logs.

Links:
47:16Episode 147

Sam Bhagwat from Mastra: the Gatsby founder building an agents framework

Sam Bhagwat is the CEO of Mastra - a typescript AI agents framework. Sam is also the cofounder of Gatsby, the popular React framework that was acquired by Netlfiy. Sam shares what he learned building Gatsby and how they're applying those lessons to Mastra. Why they're building in TypeScript, not Python. Why 20% of their users are in Japan. And why they're distributing 1,500 physical books per week on AI agents.

This episode is brought to you by WorkOS. If you're thinking about selling to enterprise customers, WorkOS can help you add enterprise features like Single Sign On and audit logs.

Links:
- Mastra
- Sam Bhagwat 
- Gatsby 
- Principles of Building AI Agents 
42:32Episode 144

Quinn Favret from Tavus: AI video API that saved our episode

Quinn Favret is the founder of Tavus. They do AI video research and products. They saved a Scaling DevTools episodes with their lipsync feature.

This episode is brought to you by WorkOS. If you're thinking about selling to enterprise customers, WorkOS can help you add enterprise features like Single Sign On and audit logs. https://workos.com/

Links:

- Tavus
- Tavus lipsync API
- Quinn Favret
- Scaling DevTools episode saved by Tavus 
50:20Episode 143

Andrew Filev, founder of Zencoder: AI Software Engineering agents

Andrew Filev is the founder of Zencoder. Zencoder is building AI coding agents.

In this episode, we explore the evolution from simple code completion AI to more sophisticated software engineering agents. While tools like GitHub Copilot revolutionized code suggestions, the next frontier involves AI agents that can handle complex engineering tasks and collaborate with each other through emerging protocols.

The discussion dives into agent-to-agent protocols, which enable AI systems to work together autonomously on software development tasks. This advancement suggests a future where AI agents could manage entire development workflows, from requirements gathering to testing and deployment. We also touch on the importance of using slower summer periods strategically - making it an ideal time for engineering teams to evaluate their tooling, processes, and prepare for upcoming development cycles.

This episode is brought to you by WorkOS. If you're thinking about selling to enterprise customers, WorkOS can help you add enterprise features like Single Sign On and audit logs.

Links
- Zencoder
- Andrew Filev
- Wrike
- Powered by Claude
- Vercel
- Perplexity AI
- Scale AI 
35:54Episode 142

Wordware founders, Filip Kozera and Robert Chandler - non-engineers can build AI workflows

In this episode we talk about Wordware, programming with LLMs, and what it now means to be a developer. Robert and Filip explain how they're building tools that let non-engineers create AI workflows, why the definition of 'developer' is changing in the AI era, and their vision for background agents that automate your work while you focus on creative tasks.

Links:
- Wordware 
- Wordware Sauna Waitlist
- Wordware is hiring
- Filip Kozera
- Robert Chandler

This episode is brought to you by WorkOS. If you're thinking about selling to enterprise customers, WorkOS can help you add enterprise features like Single Sign On and audit logs.

P.s. thanks to Oana Olteanu for making it happen  
48:48Episode 137

ChatGPT didn't kill SEO - Elston Baretto, founder of Tiiny.host

Elston Baretto is the founder of Tiiny.host - the simplest place to put your work online. In this episode we talk about how Elston has been able to grow Tiiny to 70,000+ sign ups per month with content marketing.

This episode is brought to you by WorkOS. If you're thinking about selling to enterprise customers, WorkOS can help you add enterprise features like Single Sign On and audit logs.

Links:
- Tiiny.host
- Elston Baretto
- Ramen Club
- Charlie Ward
- Sabba
- Veed 
50:29Episode 133

Sunil Pai on AI agents, Cloudflare and React

This episode is with Sunil Pai. He works at Cloudflare after his startup PartyKit was acquired. Previously he was on the React core team at Meta.

He's a great guy. And obsessed with AI agents.

This episode is brought to you by WorkOS. If you're thinking about selling to enterprise customers, WorkOS can help you add enterprise features like Single Sign On and audit logs.

Links:
- Sunil Pai on X
- Sunil Pai's site
- Building agents with Cloudflare
- PartyKit
- Durable objects 

53:52Episode 131

The startup behind ChatGPT voice - Russ d'Sa from LiveKit

Russ D’Sa is the founder of LiveKit. They are an open source tool for real time audio and video for LLM applications and they power the voice chat for ChatGPT and Character AI.

We discuss:
- How lightning works (using ChatGPT/LiveKit)
- How LiveKit started working with OpenAI
- Why Russ turned down an early 20m acquisition offer
- What it’s like to work with the fastest growing company (ever?)
- How to prepare for massive scale challenges
- Russ’s 3 letter twitter handle

This episode is brought to you by WorkOS. If you're thinking about selling to enterprise customers, WorkOS can help you add enterprise features like Single Sign-On and audit logs.

Links:
- LiveKit 
- Russ’s Twitter 
56:07Episode 123

Guillermo Rauch, founder of Vercel: Developer Experience, AI and v0

Guillermo Rauch is the founder of Vercel. Vercel is a cloud infra platform so easy to use that it’s almost become a category: “I’m building the Vercel of X”.

Vercel also recently launched v0 which is potentially the next evolution of web development - type what you want and it builds it and deploys it for you.

He’s also the creator Next.js, socket.io and a ton of other open source tools and startups. Plus he’s a prolific investor in DevTools.

I’ve missed a ton of his achievements here but essentially, he’s the king of DevTools and you probably know him already.

What we talk about
- Why Guillermo bets on people who ship
- What AI has in common with Prettier
- v0 puts design first
- Saying ‘not yet’ is a boss move
- Why Guillermo thinks devs won’t lose their jobs
- How you can learn product building
- Why you should be careful when hiring from rocketships - not everyone was in the control room
- The value of people having a full stack skill set. And why communication is more important than ever
- Why it’s so important to explain what you do in simple terms
- Tools Guillermo is excited about right now

Links:
- Guillermo Rauch
- Vercel
- v0
- NextJS
- Socket.IO
- Browserbase
- LiveKit
- Languine

This episode is brought to you by WorkOS. If you're thinking about selling to enterprise customers, WorkOS can help you add enterprise features like Single Sign On and audit logs. https://workos.com/
44:43Episode 116

Guy Podjarny, Snyk and Tessl founder - The future of programming

Guy Podjarny is the founder of Tessl - a startup that is rethinking how we build software.

Guy previously founded Snyk - a dependency scanning tool worth billions of dollars. Before Snyk, Guy founded Blaze, which he sold to Akamai.

This episode is brought to you by WorkOS. If you're thinking about selling to enterprise customers, WorkOS can help you add enterprise features like Single Sign On and audit logs.

In this conversation, we talk about the future of programming and the future of DevTools. 
  • The future of programming will focus on writing specifications.
  • Trust in AI tools
  • Snyk is an example of how tools can integrate into existing workflows.
  • Code can become disposable, allowing for flexibility in development.
  • Specifications will serve as repositories of truth in software development.
  • Developers will need to adapt their skills to leverage AI tools effectively.
  • Community collaboration is essential for the evolution of AI development tools.
  • AI simplifies and democratizes the process of software creation
Thanks to Anna Debenham for making this happen. 

38:28Episode 112

Daksh Gupta from Greptile - do marketing differently

In this conversation, Daksh Gupta, the CEO of Greptile - an AI code understanding API - shares:
  • Why it’s important to do unique types of marketing, like making an energy drink
  • Why most people misunderstand sales
  • How companies are buying AI tools and why it will probably change soon
This episode is brought to you by WorkOS. If you're thinking about selling to enterprise customers, WorkOS can help you add enterprise features like Single Sign On and audit logs.

Links:
40:01Episode 111

Ankur Goyal from Braintrust

Ankur Goyal is the founder of ​Braintrust​, a year old LLM eval platform that is already used by Figma, Vercel and Stripe and just raised $36m from a16z. It's a rocketship.

This episode is brought to you by WorkOS. If you're thinking about selling to enterprise customers, WorkOS can help you add enterprise features like Single Sign On and audit logs.

Key Success Factors
- Started with a targeted list of ~50 companies already working with AI
- Focused on early adopters and innovators in the space
- Strategy: If they could make the frontrunners happy, others would follow

Links:
- Braintrust
- Ankur Goyal
- Alana Goyal
- Basecase 
- Elad Gil 
- Martin Casado

Chapters:
* 00:00  Introduction to BrainTrust and Its Success
* 02:52  The Importance of User Research in Product Development
* 06:11  Building Relationships with Key Customers
* 09:05  The Role of Feedback in Product Improvement
* 11:54  The Impact of Mentorship on Entrepreneurial Success
* 15:11  Identifying Market Opportunities in AI Development
* 18:00  Effective User Interviews and Problem Validation
* 20:59  The Evolution of BrainTrust's Product Features
* 23:55  Advice for Aspiring DevTool Founders
* 26:48  Exciting Developments in the DevTool Space
35:28Episode 110

The story of Pydantic and Logfire | Samuel Colvin

​Samuel Colvin​ - the creator of ​Pydantic​ - the most popular data validation library for Python. Used by literally everyone (Anthropic, OpenAI, Meta, NVIDIA, even the NSA). He shares the story behind his startup ​Logfire​ which just raised $12.5m from Sequoia.

This episode is brought to you by WorkOS. If you're thinking about selling to enterprise customers, WorkOS can help you add enterprise features like Single Sign On and audit logs.

Key takeaways:
- You can just build a different product to your open source project and leverage your brand
- Quality of product matters a LOT (if you can build a popular open source project, can probably build a quality paid product)
- Really helps to be part of a movement. Hard to predict but Pydantic benefited from two (types and LLMs)
- GitHub stars are a vanity metric compared to download numbers

Links:
- Pydantic
- Logfire
- Samuel Colvin

Chapters
00:00 The Genesis of Pydantic
02:46 The Evolution of Software Development
06:02 Building a Successful Open Source Library
08:52 The Impact of Community and Adoption
11:51 Metrics of Success in Open Source
15:08 Transitioning from Pydantic to LogFire
17:59 The Vision Behind LogFire
20:50 The Connection Between Pydantic and LogFire
24:05 Navigating the Challenges of Building a Startup
26:56 The Future of Observability and Databases

P.s. thanks to my friend Abeed for making the episode happen!
42:19Episode 105

Paul Klein, CEO & Founder of Browserbase

Paul Klein is the founder and CEO of Browserbase - one of the fastest growing DevTools in 2024.

Browserbase is a headless browser API focused on helping AI Agent startups.

We dig into:
  • Why browser automation?
  • How Browserbase hit "VC-market-fit"
  • Visionary is revisionist-history 
  • Tips for hiring your friends
  • Why buying a jacket is like buying a devtool
  • Building an in-person DevTool in San Francisco
  • Making priorities (what Paul doesn’t care about).
Where to find Paul and Browserbase:
References
To support Scaling DevTools, please check out the Enterprise Ready Conf from WorkOS https://enterprise-ready.com/
76:17Episode 103

Shawn Wang (swyx) - founder of smol.ai, Latent Space, AI Engineer, DX.tips

Shawn Wang (aka swyx) is the founder of smol.ai (AI news curation), and the cohost of Latent Space (popular AI Engineer podcast).

Plus, Shawn started the AI Engineer movement with his essay Rise of the AI Engineer and organized two incredible AI engineer conferences in the past twelve months - AI Engineer World's Fair and AI Engineer Summit

And Shawn has angel invested in DevTools like Airbyte, Railway, Supabase, Replay.io, Stackblitz, Flutterflow, Fireworks.ai while running the DevTools angels community.

Besides this, Shawn curates DX.tips (DevTools magazine) and in a past life wrote the Coding Career handbook, championed learn in public, cofounded Svelte Society and was previously Head of Developer Experience at Temporal, and a Developer Advocate at AWS and Netlify.

Also, before this, Shawn had a very successful career in investment banking, trading, building data pipelines and performing quantitate portfolio management. I think this brings him a very unique perspective - I've always admired his ability to zoom out and see the big picture and the trends.

Even though Shawn is now all-in on AI, he's still one of the go-to authorities on DevTools go-to-market.

As you can tell, Shawn is someone I deeply admire. So I'm glad he came back.

What we discuss:
  • Organizing the AI Engineer Conferences
  • Rise of the AI Engineer
  • Intentionality and principles (yes we even talk about Alcoholics Anonymous)
  • The AI CEO
  • Invisible deadlines
  • Ilya believing in AGI more than most people at OpenAI
  • Are developers going to be obsolete? 
  • Thor convinced swyx to invest in Supabase
  • Building DevTools that work well with LLMs
  • Angel investing in DevTools - why and how
  • Is DevRel dead?
  • How to hire DevRel
  • Why DX.tips exists
Links:
Check out the Enterprise Ready Conf from WorkOS https://enterprise-ready.com/
39:13Episode 99

Customer support for DevTools, with Nick Gomez from InKeep

Nick Gomez is the co-founder and CEO of InKeep. InKeep is an AI customer support tool focused on Developer Tools.

They discuss the importance of understanding developer needs, the role of AI in technical support, and how community engagement can enhance support efforts.

What we discuss
  • AI support for developer tools is different from traditional B2B SaaS support.
  • Developers often seek help through documentation and community forums.
  • Scaling technical support requires understanding the developer's tech stack.
  • Clear communication channels can improve support efficiency.
  • AI solutions must prioritize quality to build trust with users.
  • Community engagement can help crowdsource support efforts.
  • Support teams should continuously improve documentation based on user inquiries.
  • 24/7 support can be achieved through AI tools.
  • Investing in customer relationships can lead to valuable insights and support.
  • Innovative tools are changing the landscape of developer support.
Links:
Keywords
AI support, developer tools, technical support, community engagement, customer investment, quality assurance, support team structure, 24/7 support, innovations in development
62:33Episode 82

Aaron Francis - how to make videos developers want to watch

Aaron Francis is someone who needs little introduction. Especially if you've ever used Laravel or MySQL.

Aaron built up the highly acclaimed PlanetScale YouTube channel and now publishes content on his own channel and founded Try Hard Studios to help developer tools make amazing video content.

Here are some quotes from Aaron's viewers:
  • hey man your videos kick ass and i cannot thank you enough for your approach with these. your videos can be watched once and understood... every single one of them... i don't know how you do it, but the way you have picked to teach anything you teach is incredible. you freaking rock! thank you!
  • Great stuff! Love that you mix in a bit of fun with the content, it's what got me to subscribe!
  • I have been working with MySQL for last 17 years and I never use cursor but your video helped me to understand MySQL cursor. Thank you
  • iterally laughing out loud several times. absolute gold.
    (partner's like "what are you watching?!" "a guy seeding a database!"
In this episode, we take a deep dive into how Aaron makes videos and what you can learn from his approach.

This episode is sponsored by WorkOS. If you're thinking about selling to enterprise customers, WorkOS can help you add enterprise features like Single Sign On and audit logs.

Links:
  • Aaron's channel: https://www.youtube.com/@aarondfrancis
  • Aaron's Twitter https://x.com/aarondfrancis
  • Mostly Technical Podcast - https://mostlytechnical.com/ 
  • Try Hard Studios: https://tryhardstudios.com/
  • Aaron's Handwriting robots - https://x.com/aarondfrancis/status/1438888219471491074?lang=en 
36:36Episode 67

OpenAI want to build the best developer product ever - OpenAI's first DevRel, Logan Kilpatrick

Guest: Logan Kilpatrick, member of OpenAI’s developer advocacy team, often described as OpenAI’s first DevRel.

Highlights:

  • Challenges and Growth: Logan discusses the evolution of developer engagement from GPT 3.5 to the explosive growth following ChatGPT's success. Initially faced with the challenge of generating developer interest, the release of ChatGPT marked a significant shift, highlighting the shift from awareness to scaling and improving developer experience amidst high demand and compute-intensive operations.
  • Developer Experience Focus: Logan emphasizes the focus on developer experience, detailing the balance between improving platform features and releasing new models and APIs. Despite past trade-offs, the goal remains to enhance core platform functionalities and developer-friendly features.
  • Decision Making and Prioritization: Logan shares insights into the dynamic and fast-paced environment at OpenAI, which requires flexibility in planning and prioritization. Key focus areas include documentation, product improvements, direct developer interactions, internal coordination, and supporting launches, especially the GPT Store.
  • Impact of Documentation: Underscoring the critical role of documentation, Logan points out that effective documentation is paramount for developer success, guiding the use of OpenAI's API and models. Efforts are underway to improve documentation quality and support various user personas beyond developers.
  • Developer Community Engagement: Lessons from engaging with the developer community include the need for diverse content formats and accommodating various user personas. Logan acknowledges the challenge of keeping documentation and resources updated in a rapidly evolving API landscape.
  • Building a Superior Developer Experience: Logan stresses the importance of OpenAI's mission to benefit everyone and the role of the API in achieving widespread impact. The commitment to providing the best tools for developers is seen as a differentiator in the competitive landscape of AI model providers.
  • Managing Attention and Feedback: Despite the challenges of being a public figure within the developer community, Logan values direct feedback for continuous improvement. Balancing public engagement with deep work, especially on documentation and launch support, is highlighted.
  • Community Questions and Answers: Logan addresses questions from the community, touching on the desire for innovative applications of OpenAI technology, plans for global events, prioritizing documentation, addressing developer concerns about scaling, and sharing personal preferences for deep dish pizza in Chicago.

Rapid Fire Community Q&A:

  • Innovative Applications: Logan hopes to see development of multiplayer, multimodal text-first AI assistants.
  • Global Events: OpenAI is expanding its presence, including hiring in London and considering events in cities like Atlanta.
  • DevRel Strategy for 2024: Focus on creating excellent documentation.
  • Developer Concerns: Addressing challenges around freedom to scale and capacity constraints.
  • Personal Time: Logan plans to take vacation during the end-of-year code freeze in 2024.
  • Chicago Deep Dish Recommendation: Lou Malnati's and Paradise Park are Logan's picks for the best deep dish pizza.

Links:
  • Logan's Twitter - https://x.com/OfficialLoganK
  • Romain's Twitter https://twitter.com/romainhuet
  • OpenAI https://platform.openai.com/
  • tlDraw https://www.tldraw.com/
  • Bloop https://bloop.ai/ 
  • Joyfill https://joyfill.io/
  • https://portkey.ai/
  • Stripe docs https://stripe.com/docs 
This episode provides a behind-the-scenes look at OpenAI's efforts to enhance developer engagement, the challenges of balancing innovation with platform stability, and the importance of community feedback in shaping the future of AI development tools.

Show notes generated with gpt4 (using a blog post I wrote) 
19:11Episode 24

Developer Marketing at a startup with Zivit Katz from Zigi

Zivit Katz is the VP of Marketing at Zigi. Zigi is an AI-powered personal assistant for developers. By managing your dev workflow and handling all your mundane, non-programming tasks across multiple apps directly from Slack.

What we cover
  • Marketing at Zigi
  • The market and users
  • Learning the right language
  • Understanding the developer
  • The importance of marketing plans
Where to hear from Zivit
13:40Episode 7

Solve problems - developer marketing with Julie Reboul

Julie Reboul is a Senior Developer Marketing Manager at Algolia. Algolia is an AI-powered search and discovery platform for dynamic experiences. Julie has also previously worked with companies such as Microsoft, Twitter, and Orange.

What we cover

(00:37): Could you tell us a bit about the kinds of things you're working on at Algolia?
(02:58): How do you cultivate a community of trust at Algolia?
(04:57): What is it that you and your partner focus on?
(06:25): What do you think attracts developers to want to join Algolia's live sessions or developer conferences?
(07:24): Could you tell us a little bit about how you approach co-marketing?
(09:10): Could you share a bit about the culture at Algolia?
(10:47): What changes have you seen in Algolia in the last five years?
(11:55): In the five years that you've been there, what do you think Algolia does well that's led to its success?

Where to hear from Julie