Y Combinator

RFS: Machine learning to simulate the physical world

Software tools relying on physics-based models for complex tasks like weather prediction and drug discovery face severe computational bottlenecks that require supercomputers and weeks of processing time. AI models function as efficient general function approximators that solve these same physics problems in seconds on standard hardware, eliminating the need for massive computational resources. This initiative targets founders seeking to enter markets that were previously infeasible by leveraging this technology to reduce prediction latency from days to seconds.

When Should You Trust Your Gut?

Dalton, Michael

Startup guidance at YC distinguishes between founders bringing deep prior expertise, who should trust their gut to build products that impress themselves, and those building expertise, who must rely on rapid learning and customer discovery to avoid over-planning. The most frequent failure mode involves role confusion where experts chase investor trends instead of leveraging their unique vision, or novices apply complex roadmaps without validating assumptions. Consequently, advisors are urged to tailor feedback dynamically, shifting from instinct-driven validation for experts to iterative market learning for generalists to align strategy with the founders' actual capabilities.

Inside The Hard Tech Startups Turning Sci-Fi Into Reality

Y Combinator advises hard tech founders to leverage a $500,000 investment and three-month timeline to validate specific technical kernels rather than building full-scale complex products. By isolating manageable milestones—such as securing massive Letters of Intent or demonstrating proof-of-concept prototypes—companies like Boom Supersonic and Relativity Space de-risk their ventures to achieve billion-dollar valuations or successful commercial launches. This strategy prioritizes "timeline compression" and tangible execution over traditional fundraising, allowing teams to mitigate technical risks while capitalizing on obvious market demand.

RFS: Better Enterprise glue

This event addresses the inefficiency of enterprise custom integration code, dubbed "dark matter," by demonstrating how large language models can automatically generate and maintain specific "glue" code for unique business systems. The discussion highlights how LLMs can disrupt the multi-billion dollar ecosystem built by vendors like Oracle, Salesforce, and NetSuite to support these manual development processes. Additionally, the presentation serves as a recruitment signal, explicitly inviting skilled professionals to apply for roles dedicated to solving this automation challenge.

How to Survive the Crypto Boom & Bust Cycle

Chandan Lodha

Cointracker co-founder Chandan Lodha guides the portfolio tracking and tax compliance platform through volatile crypto cycles by overcoming a 2020 liquidity crisis and a subsequent product obsolescence period. To regain market leadership, the company paused feature development for nine months to rebuild its offering around user obsession and long-term execution, successfully launching Cointracker 2.0 in January. Lodha emphasizes that navigating the industry's boom-bust patterns requires founders to maintain humility during bull markets and conviction during bear markets.

RFS: Climate Tech

Amidst a newly established global consensus on climate change, startups are recognized as critical drivers of solutions, having already spearheaded innovations in electric vehicles, solar energy, and carbon counting. To leverage this momentum, a curated resource outlining high-impact problems and investment opportunities has been compiled specifically to guide founders and capital toward strategic areas of growth. This actionable directory, located at ycommentary.com/RFS, provides a clear roadmap for scaling technologies like heat pumps and carbon counting to meet collective environmental goals.

How To Build Generative AI Models Like OpenAI's Sora

The event analyzes OpenAI's Sora model, highlighting its breakthrough in text rendering and temporal consistency while noting persistent flaws in physical logic and immense computational demands. It details how YC startups like Infinity AI and Sonato are achieving comparable results through data efficiency, synthetic training, and optimized model architectures rather than massive scale. The discussion concludes by projecting how these physics-simulating AI techniques will accelerate scientific discovery in fields ranging from weather prediction to drug discovery.

San Francisco (and Silicon Valley) is for the builders.

San Francisco distinguishes itself through a dense ecosystem that manufactures luck via peer collaboration, fostering a cultural environment where entrepreneurship is celebrated rather than stigmatized as an anomalous occupation. This supportive local dynamic stands in stark contrast to global norms where early-stage founders often face social ridicule, effectively validating the struggles of those building new companies. The event further contextualizes this environment by exploring how the city serves as a critical refuge for misfits, such as the speaker who migrated from Chile, to pursue legitimate and difficult startup journeys.

RFS: Real-world crypto

This initiative challenges the current crypto landscape by funding engineering teams that tokenize tangible assets like real estate, bank deposits, and actual businesses to grant enforceable ownership rights. Because legal and regulatory hurdles are paramount, the program prioritizes applicants who possess deep legal expertise or have directly collaborated with counsel to ensure compliance. Ultimately, the effort aims to shift the industry from abstract tokens to instruments backed by verified real-world value.

Why Founders Shouldn't Think Like Investors

Dalton, Michael

Dalton and Michael argue that early-stage founders trained in venture capital or corporate consulting often fail by applying large-company frameworks like extensive market analysis and exit planning to pre-product startups. This "VC mindset" creates analysis paralysis and a detachment from the critical "micro" work of manually acquiring the first customer, whereas successful execution requires a "beginner's mind" focused on deep user immersion and domain expertise. The speakers conclude that founders must unlearn these counterproductive habits to embrace the uncertainty of the zero-to-one phase and leverage the flexibility needed to navigate early business failure.

The best AI founders in the world are moving here

San Francisco has re-established its status as a global hub for artificial intelligence following the ChatGPT catalyst, which reversed the remote work exodus and concentrated major entities like OpenAI and Anthropic within a specific "Cerebral Valley" cluster. Y Combinator's strategic relocation to the Dog Patch further intensifies this agglomeration by physically co-locating hundreds of founders to maximize high-frequency networking and accelerate the current "boom loop." This renewed density is projected to drive a decade-long economic recovery, filling previously vacant downtown offices while fostering a techno-optimist society that reinvests tech wealth into housing, safety, and inclusive growth.

RFS: Commercial open source companies

Y Combinator has invested in over 150 open-source companies, adopting a strategy that leverages technical founders' comparative advantage in communicating directly with engineer users to bypass traditional sales hurdles. This approach accelerates product iteration through immediate community feedback, positioning open-source startups as emerging industry standards within their respective sectors. The organization now explicitly invites technical founders pursuing similar open-source business models to engage with their investment program.

Build product. Talk to users.

Startup founders are cautioned against engaging in high-risk activities like consuming generic advice or attending vague networking events, as these create an illusion of productivity without yielding tangible results. These behaviors are deemed most dangerous because they distract from the critical goal of building a product and acquiring users necessary to achieve product-market fit. True progress remains impossible until the business secures paying customers, rendering all other efforts worthless in the earliest stages.

Building Confidence In Yourself and Your Ideas

Dalton, Michael

This analysis warns founders against relying on superficial validation, such as cold outreach, which often leads to "pivotitis" and wasted resources driven by fear or false industry data. It emphasizes that success requires high internal conviction to withstand fundraising cycles and negative feedback, distinguishing between healthy strategic adjustments and random walks that deplete energy without generating experience. Ultimately, the event defines a true Minimum Viable Product as one that satisfies at least one user, urging teams to prioritize deep, iterative learning over rushed metrics.

We fund smart founders, irrespective of what they want to work on.

Although a prevailing public perception suggests Y Combinator prioritizes AI startups through a dedicated investment thesis, YC partners clarify that selection criteria actually center on founder intelligence rather than specific subject matter. This reality explains why approximately 50% of the Summer 2023 batch focuses on large language models, a trend partners characterize as an emergent phenomenon driven by the interests of the most qualified applicants rather than institutional bias. Consequently, the high concentration of AI ventures reflects the specific capabilities and ambitions of the founders who applied, not a strategic shift in funding priorities.