Open Source Startup Podcast

Hosted by Robby (MTF); Tim (Essence VC)

TechnologyInterviews guests

Website RSS feed

Episodes

201

Latest episode

Aug 2026

Language

About the show

The leading podcast on how to build a successful open source company. Learn from the founders of HashiCorp, Chronosphere, Vercel, MongoDB, DBT, mobile.dev and more!

Listen to episodes

60 recent

August 3, 202639 min

E201: Building Your Agent Army with Paperclip

In this episode, our co-hosts Robby and Tim talk with Paperclip Co-Founder Dotta (Nate) whose viral agent orchestration platform has taken off over the past several months as more teams adopt agents at work. Their open source, also called paperclip , has >75K stars on GitHub and is becoming the de facto way to manage a team of agents. This episode explores what it really means to build an “agent-led company.” Rather than thinking of AI agents as isolated tools, Paperclip treats them like employees working toward shared goals, with the platform acting as the company that coordinates them. We discuss why existing AI workflows (whether it’s dozens of Claude Code tabs or standalone coding agents) break down at scale, and how Paperclip provides a unified interface for managing projects, agents, tasks, context, and budgets across any model or AI harness. From marketing and documentation to customer support triage and software development, the conversation dives into how teams can assign the right agents to the right work while keeping humans in control through evaluation, reflection, and orchestration. We also unpack the challenges and opportunities of building an open source agent platform in one of AI’s fastest-moving markets. The discussion covers Paperclip’s explosive early growth, the realities of maintaining a ton of community pull requests, security risks in agentic systems, and why the future isn’t about replacing people with AI - it’s about enabling individuals and organizations to scale their impact through coordinated agents. Looking ahead, we explore Paperclip Cloud, enterprise workflows, governance, and why the next frontier for AI isn’t just better models, but better memory, context, and systems for managing an entire workforce of agents.

July 27, 202644 min

E200: Open Sourcing Warp’s Terminal

This episode has our co-hosts Robby and Tim in conversation with Warp founder Zach Lloyd as he shares the company story and decision to open source their core product: the terminal. They now provide an open source agentic development environment (born out of the terminal) which is also called warp and has almost 65K stars on GitHub. Zach discusses his frustration with the traditional terminal and how it inspired him to rebuild it from first principles. Originally conceived before ChatGPT as a more intuitive, collaborative command-line experience, Warp quickly resonated with developers and has since grown to 1 million users. Lloyd discusses the company’s decision to build Warp in Rust, its early reluctance to open source the product, and how the rise of coding agents transformed Warp from a modern terminal into a broader agentic development environment. The conversation also explores Warp’s evolving product vision: a terminal, agent harness, and cloud platform that can orchestrate different models and coding tools without locking companies into a single provider. Lloyd explains how Warp uses agents to triage issues, write and review code, test interfaces, and accelerate open source contributions while keeping humans involved in key decisions. Ultimately, the episode is about how AI is reshaping software development and Warp’s ambition to become the platform that turns ideas, community feedback, and company workflows into shipped software as efficiently as possible.

July 14, 202642 min

E199: Rethinking Infra-as-Code for the AI Era

This Open Source Startup Podcast episode has our co-hosts Robby and Tim in conversation with Pavlo Baron from Platform Engineering Labs - creators of open source infra-as-code platform formae . This podcast explores why traditional infrastructure-as-code tools like Terraform and Pulumi can struggle with state, drift, and the disconnect between code and real-world infrastructure. Pavlo shares lessons from building Instana and explains how those experiences led to Formae, an open-source approach that treats infrastructure state as accessible, executable code using Apple’s Pkl language. The conversation also looks at what infrastructure management needs in the AI era. By reducing low-level complexity and giving both humans and AI agents safer abstractions for discovering, modifying, and reconciling infrastructure, Formae aims to make infrastructure automation more reliable and less labor-intensive.

June 18, 202642 min

E198: How Unikraft Launches AI Agents in

This Open Source Startup Podcast episode has our co-hosts Robby and Tim in conversation with Dr. Felipe Huici , CEO of Unikraft - the compute layer for sandboxes, AI agents, or any workload with VM-grade isolation. Their open source, also called unikraft , has 4K stars on GitHub and provides a next-generation cloud native kernel. This episode explores how Unikraft is building infrastructure for the next generation of AI agents, arguing that agents should run in virtual machines rather than containers. The conversation focuses on the unique requirements of agentic workloads: fast startup times, the ability to pause and resume state, strong isolation, and efficient resource utilization at massive scale. Unikraft’s technology enables lightweight virtual machines that can start in under 10 milliseconds, helping companies reduce latency, lower infrastructure costs, and run large numbers of ephemeral agents on minimal hardware. The discussion also covers emerging AI infrastructure needs such as checkpointing, branching, headless browser automation, and GPU access. The podcast also traces Unikraft’s origins from an academic research project to an open-source Linux Foundation initiative and, eventually, a startup founded in 2022. The conversation examines customer adoption, the role of Unikraft as foundational infrastructure for AI platforms, competition and collaboration within the agent ecosystem, the future of GPUs and virtualization, and lessons learned from building a company in the rapidly evolving cloud and AI infrastructure market.

June 9, 202640 min

E197: The Evolution of Building Open Source Businesses from HashiCorp to Flox

This Open Source Startup Podcast episode has our co-hosts Robby and Tim in conversation with James Bayer , Chief Product Officer at software development platform Flox . Flox's open source, also called flox , provides a software environment platform powered by package manager Nix . In this episode, James shares lessons from his career across Cloud Foundry, Pivotal, and HashiCorp, where he helped turn widely adopted open-source projects like Terraform into sustainable businesses. His core takeaway is that support-only open source is difficult to scale; successful companies usually monetize the “multiplayer” capabilities that teams and enterprises need while keeping individual usage free. Now at Flox, James sees a similar opportunity built on top of Nix, a powerful but historically complex technology. He joined because Flox makes Nix dramatically easier to use, helping developers and AI agents manage software environments and dependencies. He also discussed the balance between open-source principles and commercial viability, and why he remains optimistic about the future of software development in the age of AI.

June 3, 202638 min

E196: Shifting Developer Portals to Agent Portals with Port

This Open Source Startup Podcast episode has our co-hosts Robby and Tim in conversation with Zohar Einy the Co-Founder of agentic SDLC platform Port . They have a few open source projects including ocean which allows third-party systems to integrate with their developer portal. Port is positioning itself as the infrastructure layer for the agentic era, evolving the traditional internal developer portal into what its founders describe as a “system of record for agents.” The company believes the future belongs not to vertical point solutions, but to flexible platforms that organizations control themselves, enabling anyone, from developers to non-technical employees, to become builders. Rooted in the founders’ experience of overwhelming developer workflows and ticket volumes, Port aims to centralize engineering context while making both humans and AI agents more self-sufficient. Their hybrid approach combines openness and commercial software, with public roadmaps, community contributions, and open-source integrations helping customers extend the platform while maintaining governance and control. The conversation also explored how AI is reshaping engineering organizations. Port is focused on creating the infrastructure around agents rather than building the agents themselves, providing visibility, permissions, governance, and a unified “context lake” for agent activity. As companies deploy increasing numbers of coding, security, SRE, and product agents, leaders need a control plane to understand what agents are doing and ensure they operate safely. The team is already seeing customers use Port to automate large portions of engineering support workflows, and they believe enterprises are adopting AI-driven workflows as quickly as, or faster than, mid-market companies. Internally, this pace of change requires constant adaptation, particularly across go-to-market teams, where education and flexibility have become more important than rigid playbooks.

May 26, 202645 min

E195: Taking on the New AI Attack Surface With Manifold: Runtime, Skills & Supply Chains

The latest Open Source Startup Podcast episode has our co-hosts Robby and Tim in conversation with Neal Swaelens and Oleks Yaremchuk , 2 of the Co-Founders of runtime agent security company Manifold Security . Manifold recently released Manifest , their open-access, graph-based supply chain intelligence tool for users to scan skills and plugins to uncover any potential supply chain risks. In this episode, Neal and Oleks explain why AI agents are reshaping cybersecurity - shifting the focus from guardrails to runtime security. As tools like Claude Code and Codex spread rapidly, companies often have little visibility into the agents, plugins, skills, and external assets employees are using, creating major supply chain and runtime risks. Drawing on their experience building LLMGuard and leading security teams at Protect AI and Palo Alto Networks, they argue that runtime detection and response is still a wide-open market opportunity. They also discuss what it takes to build in the crowded AI security space, where buyers now expect real products instead of roadmap promises. The conversation highlights lessons from open projects like LLMGuard and Manifest, why reducing noise and false positives matters, and how open ecosystems can help establish trust and industry standards for securing AI agents and assets.

April 29, 202641 min

E194: Fal's Bet on Generative Media

The latest Open Source Startup Podcast episode has our co-hosts Robby and Tim in conversation with Batuhan Taskaya , the founding engineer and current Head of Engineering at generative media cloud Fal . Fal is a developer platform that allows builders to develop and fine-tune models with serverless GPUs and on-demand clusters. This episode explores how a small, highly technical team carved out a unique position in the AI boom by focusing on generative media - images, video, and audio - while most of the industry rushed toward language models. Early on, they recognized that image and video models operate very differently from LLMs. With no strong API-first players in image generation, they started there and doubled down on building reliable, high-performance infrastructure for running these models in the cloud, leveraging deep expertise in systems and performance engineering. Their strategy of embracing open-source models, then fine-tuning and optimizing them for real-world use cases, helped them quickly gain traction - growing from zero to $400M of revenue by 2026 and scaling rapidly as demand for generative media surged. The conversation also dives into how the company evolved into a full-stack generative media platform, expanding from images into video and audio as those markets matured, especially with video seeing explosive growth in 2024–2025. A key differentiator has been their relentless focus on inference performance, custom kernel optimization, and cost efficiency, which has driven strong customer retention. Rather than betting on a single model, they embrace rapid model turnover and ecosystem fragmentation, ensuring flexibility for developers and enterprises alike. Looking ahead, the biggest challenges lie in scaling video models and securing enough compute capacity in a supply-constrained GPU market. Throughout, the story highlights the power of small, focused teams with clear strategy and the ability to pivot quickly in a fast-moving AI landscape.

April 8, 202639 min

E193: Managing 100s of Agents with Maestro

In our latest Open Source Startup Podcast episode, co-hosts Robby and Tim talk with Pedram Amini , the creator of open source platform Maestro which allows users to run fleets of AI coding agents autonomously for long periods of time. Their project has 3K stars on GitHub. The episode explores how Maestro's multi-agent system overcame a key limitation in generative AI: context overload. After juggling many Claude sessions for different tasks, Pedram realized each problem needed its own isolated workflow. Maestro turns this into a system letting users run many agents and tabs in parallel, keeping tasks separate and avoiding context degradation during long or complex work. Maestro is designed for scale, enabling dozens or even hundreds of agents to handle complex projects simultaneously. It’s flexible, model-agnostic, and especially useful for breaking big problems into independent units. The project has quickly grown into a community-driven effort, reflecting a broader shift: instead of buying a bunch of tools, developers can build highly customized AI systems themselves, pointing toward a future of large-scale agent orchestration.

February 18, 202641 min

E192: Creating Browser Use, Navigating Hyper Growth & Building in the Competitive Browser Automation Space

In our latest Open Source Startup Podcast episode, co-hosts Robby and Tim talk with Magnus Müller , the Co-Founder & CEO of Browser Use - the platform that makes web agents come to life. Their open source, browser-use , has almost 80K stars on GitHub and is widely adopted. This episode dives into the unexpected rise of an open-source browser automation project that took off during Y Combinator - while many similar projects before and after it never gained traction. The founder reflects on why: delivering a “magical moment” fast. Early demos showing AI controlling a browser, inspired by trends like OpenAI’s Operator, and immediately clicked with people. What began as a developer-only Python library evolved into a hosted product as non-technical users - from sales teams to startups - wanted access. Along the way, the team leaned into controversial but compelling use cases, like AI applying for jobs on your behalf, which sparked conversation and accelerated growth. The core challenge they focused on solving was reliability: unlike deterministic automation scripts, AI agents can behave unpredictably, making trust and repeatability central problems to overcome. The long-term vision goes beyond UI automation toward agents that can skip the browser entirely and interact directly with website servers through structured actions. But the conversation isn’t just about infrastructure. The founder admits that early growth came mostly from building and talking to users, while recent months have been dedicated to storytelling and marketing rather than coding. A personal through-line emerges as well: learning to replace defensiveness with curiosity - questioning assumptions, staying open to feedback, and continuously refining both the technology and the narrative around it.