Latest Interviews

MIT 6.S094: Deep Learning for Human-Centered Semi-Autonomous Vehicles

Researchers are collecting billions of high-speed video frames from semi-autonomous Teslas to train deep learning models that detect critical driver metrics such as body pose, gaze direction, and cognitive load. By shifting from fully supervised to semi-supervised annotation strategies, the team achieves an 84-fold reduction in human effort while accurately identifying micro-saccades and emotional cues to overcome current privacy and trust barriers. This data-driven approach aims to replace static crash test assumptions with dynamic occupant monitoring, ultimately enabling vehicles to adapt passive safety systems based on real-time human behavior.

Office Hours with Kat Manalac

Kat Manalac, Craig Cannon, Kat Mignolet

Y Combinator partner Kat Mignolet outlines critical strategies for global incubators, emphasizing wealth reinvestment models and local mentorship recruitment while detailing the firm's new transparent recommendation system. She advises founders to prioritize deep validation with 100 users over broad adoption, formalize co-founder equity agreements immediately, and leverage warm introductions for press coverage. The discussion further clarifies YC's fixed $120,000-for-7% equity deal structure and encourages failed alumni to reapply as their evolution is often recognized by investors.

Sam Altman's Whale AMA

Sam Altman

The speaker outlines a strategy for high-growth entrepreneurship rooted in Paul Graham's philosophy, asserting that high school students can successfully launch startups while attending school and that mastery requires a decade of constant collaboration. Looking ahead, the event details Y Combinator's plans to expand its influence through a new online course, funding for non-profits like the ACLU, and a specific focus on addressing economic inequality caused by automation. Additionally, the discussion covers practical operational advice, including the use of Stripe Atlas for international founders, personal productivity tactics involving a vegetarian diet, and the persistence needed to navigate enterprise sales cycles.

Lessons From Doing YC Twice - Harj Taggar

Harj Taggar

Technical hiring marketplace TripleByte leverages a skill-based screening process for 15,000 engineers to match candidates with companies based on product-focused proficiency and cultural fit rather than traditional resume credentials. Founder Harj Taggar applies insights from his Y Combinator experience to advocate for live coding evaluations and deep domain expertise as primary filters for early-stage validation and successful fundraising. By prioritizing organic co-founder relationships and data-driven growth metrics over theoretical backgrounds, the platform addresses the shifting market demand for engineers who drive user growth and revenue impact.

MIT 6.S094: Recurrent Neural Networks for Steering Through Time

The lecture provided a comprehensive technical overview of recurrent neural networks, contrasting vanilla architectures with Long Short-Term Memory (LSTM) units to address vanishing gradient challenges in processing sequential data. It detailed core optimization mechanics, including backpropagation and gradient stabilization, while showcasing diverse applications ranging from machine translation and medical diagnosis to autonomous driving systems that utilize image sequences to predict steering and speed. The session concluded by emphasizing the heavy reliance on manual hyperparameter tuning and massive datasets, while setting the stage for future discussions on driver state analysis and an upcoming White House AI policy speaker.

Talks at GS - Adena Friedman: The Future of FinTech

Adena Friedman

Nasdaq executive Dina Dubrow outlines how cloud infrastructure and machine intelligence are reshaping the equities exchange's technology landscape while blockchain promises to revolutionize settlement efficiency. The platform also launched the Nasdaq Private Market to facilitate liquidity for private companies navigating modern shareholder limits and extended lifecycles before potential public returns. Dubrow further emphasizes a leadership philosophy that balances strategic data with intuitive risk-taking, asserting that organizations and individuals must accept opportunities to learn and grow despite incomplete information.

MIT 6.S094: Convolutional Neural Networks for End-to-End Learning of the Driving Task

This lecture explores the application of Convolutional Neural Networks to computer vision challenges, specifically demonstrating how deep learning models surpass human performance on benchmarks like the CIFAR-10 dataset to enable autonomous driving systems. It details the architectural differences between convolutional, pooling, and fully connected layers while contrasting browser-based training tools like ConvNet.js with robust offline implementations in TensorFlow. The session concludes by addressing critical industry hurdles such as data scarcity for rare edge cases and the necessity for near-perfect accuracy to ensure safety in real-world deployment scenarios.

10,000 Small Businesses UK Participants Share Experiences of Time on Programme

Launched in 2012, this intensive program guided struggling entrepreneurs through a rigorous diagnostic process that identified critical financial leaks and strategic pivots from B2B to B2C models. By pairing founders with growth experts and fostering a transparent cohort of vetted peers, participants transformed their ventures from loss-making entities into cash-positive businesses generating up to £1.5 million in revenue. The initiative delivered a transformative return on investment, enabling companies to hire strategically, secure major capital projects, and establish a lasting support network that continues to drive long-term operational efficiency.

MIT 6.S094: Deep Reinforcement Learning for Motion Planning

Lex

Participants in the Deep Traffic competition develop deep reinforcement learning agents to navigate a seven-lane highway simulation, aiming to achieve an average speed of 65 mph or higher through autonomous decision-making. Utilizing client-side JavaScript and Andrej Karpathy's ConvNet.js library, competitors train neural networks via experience replay and Q-learning algorithms without relying on explicit ground truth data for vehicle actions. The initiative serves as a practical framework for exploring the complexities of autonomous driving, highlighting both the potential of simulation-based learning and the critical challenges of aligning reward functions with real-world safety constraints.

Jeff Immelt: The Transformation to a Digital Industrial Company - Talks at GS

Jeff Immelt

Under Jeff Immelt's leadership, General Electric transformed from a centralized entity into a decentralized organization leveraging its $2 trillion installed asset base to drive the Industrial Internet strategy. Facing stagnant industrial productivity, the company now competes against tech giants by prioritizing data-driven solutions and balancing active listening with decisive action. This approach supports a globalization model focused on local market access rather than wage arbitrage, attracting top talent through a meritocratic culture that empowers young leaders to manage major engineering projects.

The Agenda: Minouche Shafik discusses education and inequality

Minouche Shafik

Drawing on personal experience from post-revolution Egypt and global economic trends, the speaker argues that nationalization and rigid educational systems have eroded social mobility while lifelong learning gaps disproportionately favor the wealthy. The presentation critiques the failure of traditional models, including German vocational training, to prepare workers for careers that may shift two to five times, a disconnect that fuels public backlash against elitist experts. Consequently, the argument posits that policy reform depends on replacing jargon with accessible communication to bridge the divide between economic realities and public perception.

Populism is reshaping our world

Matteo Renzi, Beppe Grillo, Jeff Siegel, Virginia Raggi, Robert Callista, Russell Hancock, Harry Part Ovi, Sebastian Thrun

Following the 2016 Italian referendum defeat of Prime Minister Matteo Renzi, the Five Star Movement capitalized on economic stagnation to emerge as the primary opposition party, signaling a broader global surge in populism driven by automation-induced inequality. While experts identify technological displacement, particularly in sectors like trucking, as a root cause of economic dislocation, leaders such as Sebastian Thrun and Harry Part Ovi advocate for systemic reforms including lifelong learning and accessible retraining programs. Ultimately, restoring stability to Western democracies requires political and business leaders to address the grievances of alienated populations through concrete economic solutions rather than allowing technological disruption to fuel further anti-establishment sentiment.

MIT 6.S094: Introduction to Deep Learning and Self-Driving Cars

Lex Friedman, Dan Brown, William Angio, Spencer Dodd, Benedict Jenick, Andrej Karpathy, Hans Moraveck

MIT Course 6S094, led by Lex Friedman, utilizes self-driving cars as a case study to teach deep learning through two simulation projects: the reinforcement learning game Deep Traffic and the image-based control system Deep Tesla. The curriculum contrasts standard supervised learning with complex real-world challenges such as adversarial attacks and data inefficiency, requiring students to train neural networks to drive virtual vehicles at speeds exceeding 65 mph for credit. By analyzing the architectural modules of autonomy and historical milestones like the DARPA Grand Challenge, the course bridges theoretical computer science with the practical safety constraints of deploying artificial intelligence in unstructured environments.

Office Hours with Sam Altman

Sam Altman, Craig Cannon

This session outlines a B2B customer acquisition strategy where young founders leverage YC networks and peer-to-peer outreach to secure initial clients, while noting that non-technical leaders must compensate with product design and engineering proficiency. Discussion also covers critical career decisions, emphasizing the need for pre-deal operational autonomy during acquisitions and highlighting emerging high-impact sectors like AI, clean energy, and spatial computing as prime targets for new ventures. Additionally, the dialogue examines the limitations of equity crowdfunding and the specific advantages youth founders possess, such as stamina and risk tolerance, alongside the necessity of recruiting experienced advisors to balance operational gaps.

Q&A with YC Partners at Startup School SV 2016

Kat, Adora, Tim, Jaredy, Michael, Dalton, Kirsty

This event delivers a comprehensive strategic guide for startup founders, covering critical phases from immediate idea validation and co-founder selection to operational tooling and the YC application process. Key insights emphasize prioritizing genuine passion over impressiveness, balancing product perfection with growth based on specific retention thresholds, and leveraging community support rather than isolating oneself. The session concludes with logistical details for a reception in the sunken garden, offering a direct forum for continued dialogue with the YC partners and staff.