Y Combinator

The AI Agent Economy Is Here

Founders and non-technical CEOs are rapidly adopting autonomous AI workflows, creating a parallel "agent economy" where software selection is driven by swarm intelligence rather than human developers. This shift forces a strategic pivot for developers, requiring tools to offer superior, structured documentation and open APIs to be chosen by autonomous agents as primary decision-makers. Industry leaders like Maltbook and Supabase are capitalizing on this transition, while institutions like Y Combinator begin redefining their frameworks to support a future where agents act as the primary economic actors choosing infrastructure.

What Boris Cherny Learned From Building Claude Code

Boris Cherny

Quad Code's founder executed a grueling three-month sprint to launch a terminal-based product, accepting current code-generation limitations as an inevitable starting point rather than a final state. This strategy mirrors Anthropic's philosophy of building software for capabilities expected six months into the future, encouraging founders to target where models currently fail rather than optimizing for today's constraints. By prioritizing this forward-looking approach, both entities anticipate that natural model progression will resolve existing gaps without requiring premature adaptation to immature technology.

Inside Claude Code With Its Creator Boris Cherny

Boris Cherny, Mark Mandelmann, Jr., Francesc Campoy, Melanie Warrick

Anthropic has accelerated its engineering velocity to a 150% growth rate by leveraging *Claude Code* to generate 70% to 90% of its own source code and rewriting its entire codebase every six months to avoid technical debt. The organization operates a bimodal workforce of specialists and generalists while predicting that traditional software engineering roles will fade as AI automates coding tasks. With *Claude Code* now powering the majority of developer sessions and influencing 4% of public industry commits, the company is simultaneously expanding into non-technical domains through *Co-work* and monitoring its safety rating at ASL 3.

The New Way To Build A Startup

Garry

Companies such as GigaML, Legion Health, and PhaseShift are redefining operational scale by automating all internal workflows—from code integration to clinical operations—rather than merely augmenting isolated functions. This "20X company" model enables lean teams of just a few employees to service Fortune 500 clients and achieve fourfold revenue growth while maintaining flat headcounts that would traditionally require hundreds of staff. By treating internal processes as the primary product, these organizations achieve AGI-level productivity, allowing them to compete directly with massive incumbents without the corresponding payroll burdens.

Why OpenClaw Broke The Internet

OpenClaw distinguishes itself as a local AI agent that executes autonomously on user hardware with full permissions to interface directly with IoT devices and control physical environments. By leveraging comprehensive access to a user's entire local dataset, the system successfully reconstructed detailed historical narratives from forgotten files to deliver deep, unexpected insights. Unlike cloud-based competitors, this platform grants users granular control over their digital footprint while enabling direct interaction with tangible assets such as vehicles, appliances, and smart beds.

AI Native Hedge Funds

A former quantitative researcher identifies the current inflection point where AI-native strategies are displacing legacy quantitative trading models, noting that established hedge funds are already lagging due to slow adoption. The initiative targets the development of autonomous "swarms of cloud agents" capable of synthesizing financial data and executing trades without human intervention to secure a decisive competitive advantage. This speaker is currently recruiting through Y Combinator to build the first fully automated financial firm that mirrors the paradigm shifts seen in the 1980s.

Stablecoin Financial Services

The proposed Genius and Clarity Acts position compliant crypto-native stablecoins in a unique regulatory niche that blends the efficiency of DeFi with the compliance standards of traditional finance. This framework aims to resolve the current market binary by enabling yield-bearing accounts, tokenized real-world asset access, and cost-effective cross-border payments within established legal boundaries. As regulators open this window for infrastructure development, the event highlights specific opportunities for builders to scale these hybrid financial solutions through accelerators like Y Combinator.

AI for Government

Despite the private sector's rapid adoption of AI for form processing, government agencies still rely on manual systems, creating a critical need for automated solutions to manage surging application volumes. This event addresses the unique challenges and high-value rewards of penetrating the public sector, where successful vendors gain sticky long-term contracts that can expand significantly after initial acquisition. The speaker specifically invites skilled developers to join the effort of building specialized AI tools that bridge this digital divide and replicate global efficiency benchmarks within government operations.

OpenClaw Creator: Why 80% Of Apps Will Disappear

Raphael Schaad, Peter Steinberger

Launched in 2025, the open-source OpenClaw project has rapidly accumulated over 160,000 GitHub stars by introducing a local-executing personal AI agent capable of autonomously controlling hardware and synthesizing private data. Creator of the "swarm intelligence" philosophy, the project distinguishes itself through a unique security architecture that isolates user memory in local markdown files and employs a proprietary `soul.md` file to maintain core personality traits against external manipulation. As the technology evolves toward multi-agent negotiations and the obsolescence of traditional applications, OpenClaw positions itself as a value-retention model where specialized bots manage distinct domains while the primary asset remains the user's locally stored data memory.

Infrastructure for Government Fraud Hunters

A new venture proposes replacing obsolete, rule-based government fraud detection systems with an AI platform capable of automatically parsing unstructured medical records to generate lawyer-ready case files for the False Claims Act. Targeting whistleblower law firms and inspector generals, the company seeks to replace manual, years-long investigations with a solution that accelerates fraud recovery tenfold. Founded by experts in healthcare compliance or legal counsel, this initiative aims to recover billions in taxpayer losses while capitalizing on current bipartisan political momentum for fiscal accountability.

We're All Addicted To Claude Code

Calvin French-Owen

Industry analysis indicates a decisive shift toward Command Line Interfaces as the primary interaction layer for coding agents, with Anthropic's Cloud Code utilizing subagent models and OpenAI's Codex employing periodic compaction to manage context windows. Senior engineers are increasingly adopting a managerial role to direct autonomous workflows, while top-tier practitioners distinguish themselves by enforcing rigorous testing protocols and minimizing codebase noise to prevent model degradation. This evolution fosters a tension between rapid, bottom-up grassroots adoption driven by speed and enterprise-mandated top-down security controls, fundamentally reshaping software development structures toward smaller teams operating under vast agent armories.

AI Guidance for Physical Work

A new technological framework deploys real-time AI guidance via wearable cameras to accelerate physical labor proficiency, effectively replacing traditional months-long training with immediate visual and auditory coaching. Market viability is driven by the convergence of advanced multimodal models, ubiquitous smart hardware, and urgent skilled labor shortages, enabling three distinct business models: enterprise B2B sales, vertical integration in specific sectors, or a platform for rapid workforce upskilling. The initiative seeks collaborators to build systems that provide AI superpowers to field service, manufacturing, and healthcare workers, aiming to augment rather than replace human capability in physical roles.

Cursor for Product Managers

Identifying a critical gap where current AI tools like Cursor excel at implementation but fail to support the product discovery phase, the event proposes an "AI-native system" designed as a dedicated "cursor for product management." This proposed solution ingests customer interviews and usage data to generate validated feature outlines, propose specific UI and data model changes, and break down development tasks for coding agents. The organizers are actively seeking engagement from founders and developers to collaborate on building solutions that address this evolving need in product specification methodology.

Why Your Startup Website Isn't Converting

Jorn van Dijk, Aaron Epstein

Jorn van Dijk conducted a comprehensive usability audit of five AI startups—Lumari, Ally, Juicebox, Leaping AI, and The Hog—identifying critical friction points where visual complexity and premature calls to action hindered user understanding. The analysis revealed a consistent pattern of missing product proof, such as abstract illustrations or generic templates, which prevented prospects from validating value propositions before encountering mandatory sign-up walls. Recommendations focus on replacing decorative animations with functional UI demonstrations, clarifying enterprise-level scope through targeted visuals, and streamlining messaging to ensure immediate alignment between brand identity and the specific B2B solutions offered.

Building The World's First Reusable Rockets

Andy Lapsa, Tom Feldman

Recognizing a critical industry flaw where single-use rockets are discarded after a single flight, founders launched a venture dedicated to developing rapidly reusable orbital vehicles to fill this technological vacuum. They proceeded with the conviction that full reusability was a necessary pursuit, accepting the venture as the "easiest hardest decision" based on a moral imperative to attempt the solution regardless of the probability of failure. This bold entry into the aerospace market was driven by the belief that the potential for a sustainable launch model justified the risks of building a new company from scratch.