How to Build Artificial Superintelligence | Jonathan Siddharth, Founder & CEO of Turing

The case for context graphs | Aaron Levie (Co-founder & CEO, Box)

Aaron Levie has been on the podcast twice before. After we published our context graphs thesis, he wrote a response - so we invited him on to continue the conversation. A context graph is institutional memory for how an organization actually makes decisions: not how the process doc says it should, but how it works in practice. Enterprise software is very good at recording outcomes - the final price, the approved discount, the escalated ticket - but not the reasoning behind them. Which exceptions applied? What precedent mattered? Who approved what, and why? We call these missing records decision traces. Over time, they accumulate into a context graph: a living, queryable map of how an enterprise actually makes decisions, stitched across systems and time so precedent becomes searchable. We think the companies that capture that layer will define the next generation of enterprise software. Aaron read the piece and joined us to push it further. We get into how the services as software opportunity unfolds as agents scale, and what it actually takes to move them out of the sandbox and into production. Chapters: * 00:00 Intro: Aaron’s third time on the podcast! * 00:43 What is a context graph? * 01:35 Aaron’s take: this won’t be zero-sum * 04:10 Why systems of record may become more valuable in a world with 100x more agents * 05:26 The difference between data and context * 06:28 The moat incumbents have: workflow wiring, permissions, access controls * 10:35 Which functions are most vulnerable to disruption * 15:43 The trillion-dollar greenfield: high headcount, exception-heavy workflows * 16:20 Ops as a wedge: RevOps, DevOps, SecOps, and glue work between systems * 19:27 How PlayerZero is building context graphs for real engineering workflows * 20:48 Is permissioned inference possible? * 21:27 “Agents can’t keep secrets”: why access controls are so important * 27:45 What Box is building * 29:33 Multimodal context: screenshots, audio, and video * 32:50 Why vibe coding won’t change the software industry’s power-law * 35:03 Aaron’s advice to founders

Why context graphs are the missing layer for AI

My guests today are Animesh Koratana and Jamin Ball.  Animesh is the founder and CEO of our portfolio company PlayerZero, which is building AI production engineers that operate complex enterprise software autonomously - resolving production incidents, catching defects before release, and building durable models of how systems actually behave. Jamin is a partner at Altimeter Capital and the writer behind Clouded Judgement, a Substack where he analyzes emerging trends in enterprise software.  Jamin recently sparked a debate with an essay titled “Long Live Systems of Record.”  His core argument is that while agents are changing how software is used and where value accrues, they still depend on ground truth. Systems of record won't disappear so much as get pushed down the stack as new agent-native interfaces emerge on top. My partner Jaya and I felt compelled to respond, with Animesh contributing insights based on what he's seeing on the ground as he builds PlayerZero.  From our perspective, the missing layer is what happens inside the workflow itself: the judgment, exceptions, and reasoning that agents and humans apply as work gets done. We call these decision traces, and we believe the context graph they form over time will become the most valuable asset for companies building and deploying AI systems. It's a genuine debate - and one that's only going to matter more as agents move from demos to production. Looking forward to keeping the conversation going! Chapters * 00:00 Why Jamin’s essay sparked debate  * 00:35 Jamin’s argument: agents need ground truth  * 02:00 Animesh on why context graphs matter  * 07:58 What today's systems of record are missing  * 08:28 How PlayerZero thinks about context graphs  * 10:00 How context graphs could change org structures  * 11:10 How do you capture decisions without making people log everything?  * 14:35 Which systems of record are most at risk  * 17:04 Two workflows ripe for disruption: GTM and software development  * 22:31 Animesh on where context graphs can add most value  * 28:50 Why context graphs create moats for startups  * 30:00 Will context graphs be industry-specific or universal?  * 34:00 Bear case: do context graphs fail like semantic layers?  * 43:27 2026 predictions: big AI IPOs, world models, enterprise agent adoption  * 45:00 Hot takes: point solutions die; AI job-loss discourse hits a fever pitch  * 47:30 Jevons paradox: why agents create more work, not less

How to Build Artificial Superintelligence | Jonathan Siddharth, Founder & CEO of Turing

My guest today is Jonathan Siddharth, co-founder and CEO of Turing. Jonathan incubated Turing in Foundation Capital’s Palo Alto office in 2018. Since then, it has grown into a multi-billion dollar company that powers nearly every frontier AI lab: OpenAI, Anthropic, Google, Meta, Microsoft, and others. If you’ve seen a breakthrough in how AI reasons or codes, odds are Turing had a hand in it. Jonathan has a provocative thesis: within three years, every white-collar job, including the CEO’s, will be automated. In this episode, we talk about what it will take to reach artificial superintelligence, why this goal matters, and how the agentic era will fundamentally reshape work. We also dig into his founder journey: what he learned from his first startup Rover, how he built Turing from day one, and how his leadership style has evolved to emphasize speed, intensity, and staying in the details. Jonathan has been at the edge of AI for years, and he has the rare ability to translate what’s happening at the frontier into lessons for builders today. Hope you enjoy the conversation!  Chapters:  * 00:00 Cold open * 00:02:06 Jonathan’s backstory: his experience at Stanford * 00:06:37 Lessons from Rover * 00:08:39 Early Turing: incubation at Foundation Capital and finding PMF * 00:13:52 Why Turing took off * 00:15:12 Evolving from developer cloud to AGI partner for frontier labs * 00:16:49 How coding improved reasoning - and why Turing became essential * 00:20:38 Founder lessons: building org speed and intensity * 00:23:33 Why work-life balance is a false dichotomy * 00:24:17 Daily standups, flat orgs, and Formula One culture * 00:25:15 Confrontational energy and Frank Slootman’s influence * 00:29:50 Positioning Turing as “Switzerland” in the AI arms race * 00:34:32 The four pillars of superintelligence: multimodality, reasoning, tool use, coding * 00:37:39 From copilots to agents: the 100x improvement * 00:40:00 Why enterprise hasn’t had its “ChatGPT moment” yet * 00:43:09 Jonathan’s thoughts on RL gyms, algorithmic techniques, and evals * 00:46:32 The blurring line between model providers and AI apps * 00:47:35 Why defensibility depends on proprietary data and evals * 00:55:20 RL gyms: how enterprises train agents in simulated environments * 00:57:39 Underhyped: $30T of white-collar work will be automated

How to Turn Research Into Real Companies | Ion Stoica, Co-founder and Executive Chairman, Databricks

My guest today is Ion Stoica, professor of computer science at UC Berkeley and the co-founder of Conviva, Databricks, and Anyscale. Over the last two decades, Ion’s research labs - the AMP Lab, the RISE Lab, and now the Sky Computing Lab - have seeded a generation of category-defining companies.  Ion has the unique ability to turn non-consensus ideas into durable businesses. He applied machine learning to video optimization with Conviva before AI became mainstream. He scaled Apache Spark into a $60B platform with Databricks. And now, with Anyscale, he’s betting on Ray as the foundation for distributed AI workloads.  In this episode, we dig into both sides of Ion’s work: how to build world-class research labs, and how to turn research into real companies. His clarity of thought makes the future feel legible, and his track record suggests he’s very often right.  Hope you enjoy the conversation!  Chapters:  * 00:00 [https://www.youtube.com/watch?v=LyAOmClw6VE] The Spark thesis: win the ecosystem first, monetize later  * 01:00 [https://www.youtube.com/watch?v=LyAOmClw6VE&t=60s] Intro: From lab to company - Ion’s repeatable playbook  * 03:00 [https://www.youtube.com/watch?v=LyAOmClw6VE&t=180s] Did you always plan to become a founder, or did it just happen?  * 05:23 [https://www.youtube.com/watch?v=LyAOmClw6VE&t=323s] Let’s start with Spark - how did the project come about?  * 13:04 [https://www.youtube.com/watch?v=LyAOmClw6VE&t=784s] What were the most important early decisions at Databricks?  * 23:49 [https://www.youtube.com/watch?v=LyAOmClw6VE&t=1429s] You were the first CEO - what did you have to learn (or unlearn)?  * 30:01 [https://www.youtube.com/watch?v=LyAOmClw6VE&t=1801s] How was building Anyscale different from building Databricks?  * 33:53 [https://www.youtube.com/watch?v=LyAOmClw6VE&t=2033s] What’s obvious to you about the future of AI that others miss?  * 37:31 [https://www.youtube.com/watch?v=LyAOmClw6VE&t=2251s] Why AI works so well for code  * 41:00 [https://www.youtube.com/watch?v=LyAOmClw6VE&t=2460s] The thesis behind OPAQUE Systems  * 44:06 [https://www.youtube.com/watch?v=LyAOmClw6VE&t=2646s] Future infra will be heterogeneous, distributed, and vertically integrated  * 49:03 [https://www.youtube.com/watch?v=LyAOmClw6VE&t=2943s] China’s edge: faster diffusion from lab to market  * 53:19 [https://www.youtube.com/watch?v=LyAOmClw6VE&t=3199s] Platform companies still work, but only with the right investors  * 55:57 [https://www.youtube.com/watch?v=LyAOmClw6VE&t=3357s] What role did the Databricks Unit (DBU) play in value capture?  * 58:02 [https://www.youtube.com/watch?v=LyAOmClw6VE&t=3482s] AI progress is plateauing, but adoption is just beginning

How to Lead with Empathy and Resilience (Ramesh Srinivasan, Senior Partner, McKinsey)

My guest today is Ramesh Srinivasan, a senior partner at McKinsey and trusted advisor to some of the world’s top CEOs. Over his career, Ramesh has worked with leaders at companies like Cognizant, Moderna, Nissan, and Delta, helping them navigate tough challenges and scale high-performing teams. Ramesh just published a new book, The Journey of Leadership, which distills lessons from thousands of hours spent alongside top executives. In our conversation, he shares practical insights for founders on how to discover their natural leadership style, why empathy is a non-negotiable leadership skill, and what it really takes to inspire people at scale. Hope you find this conversation valuable! Chapters:  * 00:00:00 Cold open * 00:01:14 Ramesh’s backstory * 00:04:34 Things that shaped Ramesh’s leadership philosophy * 00:05:19 The big idea behind his book: The Journey of Leadership * 00:09:12 Building empathy: For your team, your customers, your market * 00:11:00 Lessons from Frank D’Souza at Cognizant * 00:14:20 Lessons from Stéphane Bancel at Moderna * 00:17:15 Trust, vulnerability, and the power of asking for help * 00:19:40 Finding purpose: Starting from life’s crucible moments * 00:22:00 Renewal: How great leaders evolve over time * 00:24:00 Common mistakes founders make on the leadership journey * 00:26:10 Resilience: The ultimate test of a founder’s staying power * 00:30:20 The impact of AI on leadership and organizational change * 00:34:00 Where AI is reshaping healthcare today * 00:36:00 Advice for AI + healthcare founders