Interview, Explained

In-depth breakdowns of tech interviews — section-by-section analysis with diagrams, quotes, and insights.

~55 min
a16z Podcast·

Travis Kalanick on Why Physical Industries Become Atoms-Based Computers

Travis Kalanick· Founder

Travis Kalanick maps the physical world onto computing — manufacturing, real estate, and logistics as the CPU, storage, and network of “atoms-based computers” — and argues food, mining, and transport are trillion-dollar industries about to be automated in a second industrial revolution.

AI InfrastructureRoboticsBusiness StrategyAI Company
~44 min
a16z Podcast·

McCall & Schmidt on Sales: Execution Beats the Playbook

Andy McCall and Joe Schmidt· Go-to-market partners

Two a16z go-to-market partners lay out the two enterprise sales playbooks for AI startups — win a few marquee lighthouse logos where proof travels, or land-grab the mid-market on ROI math — and argue which one you pick matters far less than how fast you execute it.

Business StrategyAI CompanyAgents
~46 min
No Priors·

Erik Allebest on chess thriving after machines beat it

Erik Allebest· Co-founder and CEO

Chess.com's founder bootstrapped a roughly $200M business from a domain investors dismissed as uninvestable, and argues that superhuman machines make people value human skill more, not less.

AI CompanyBusiness StrategyReasoning
~29 min
Equity·

Mina Fahmi on wearables that extend you, not replace you

Mina Fahmi· Co-founder and CEO

Sandbar CEO Mina Fahmi argues the wave of always-on AI wearables gets the relationship backwards: his Stream voice ring is push-to-talk by design, because a device only earns a place on your body when it feels like a controllable extension of you rather than a passive recorder listening in.

MultimodalAI CompanyBusiness Strategy
~56 min
TWIML AI Podcast·

Fatih Porikli on making image models precise, not just plausible

Fatih Porikli· VP of Technology

Qualcomm's Fatih Porikli argues that image generation's next frontier isn't a bigger model but smarter decomposition: reward the right objective, split planning from rendering, route to specialists, and tile the work in latent space so precise, high-resolution images run on a phone.

MultimodalInferenceAI Infrastructure
~51 min
a16z Podcast·

Garry Tan on why a markdown file is now an employee

Garry Tan· President and CEO

Garry Tan argues agentic coding rewrites startup economics: a skill file becomes a tireless employee, one operator can outwork a department, and a world built for memory-limited humans is suddenly up for grabs — just slower than the hype suggests.

AgentsAI CompanyBusiness Strategy
~15 min
Full Stack·

Ali Haghani on why AI agents need a company brain

Ali Haghani· Co-founder & CEO

Circleback's co-founder argues that as AI agents do more work, conversation context becomes scarce infrastructure: capture broadly, share it only with the right people, and reserve consequential judgment for humans.

AgentsAI CompanyBusiness Strategy
~95 min
Latent Space·

Chai Discovery on turning drug discovery into an engineering discipline

Matt McPartlon & Neil Patil· Co-founder & product lead

Chai Discovery's research and product leads explain how folding and design models turned antibody discovery from blind trial-and-error into declarative precision engineering — and why they sell the models, not the drugs.

AI CompanyTrainingReasoningOpen Source
~95 min
Latent Space·

Matt McPartland & Neil Patil on turning biology into a software factory

Matt McPartland & Neil Patil· Co-founder & Product Lead

Chai Discovery's founders explain how AI structure and design models are turning antibody discovery from an obscure, trial-and-error natural science into a software-like precision-engineering discipline.

LLMTrainingAI CompanyAI Infrastructure
~133 min
Dwarkesh Patel·

Ryan Greenblatt on why recursive self-improvement is plausible

Ryan Greenblatt· Chief Scientist

Redwood Research's Ryan Greenblatt makes the case that automating AI R&D could compress four or five years of progress into a single year — then argues that the same speed would raise the risk of catastrophic misalignment, because AIs trained to chase a high score can learn to cheat, cover it up, and possibly take over rather than turn evil.

AI SafetyReasoningTrainingAI Infrastructure
~24 min
Power Players·

Brian Chesky on Why AI Is the Best Thing to Happen to Airbnb

Brian Chesky· Co-founder & CEO

On Airbnb's earnings day, Chesky reframes AI as a tailwind rather than a threat: it helped re-accelerate revenue growth from 10% to 17% and ship nearly twice the features, while he bets the real opening is the under-built consumer side of AI and sees no near-term path for chatbots to become the place people actually book travel.

AI CompanyBusiness Strategy
~42 min
Y Combinator·

Peter Steinberger on why fun is velocity and your name is the moat

Peter Steinberger· Creator

The creator of OpenClaw retraces eight months from a hungry-night hack to a viral open-source agent that drew 18,000 issue and PR authors, and pulls out the durable lessons: annoyance finds the gaps, your name outlasts any product, a dependency's business model becomes yours, and fun is the real velocity.

AgentsAI CompanyOpen Source