Advertise on podcast: Inside The Silicon Mind

For eight years, his company sold document review. The product that finally worked came from a one-page internal tool built for his own team, one that customers saw at a conference and immediately asked to buy. I sat down with Nischal Nadhamuni, co-founder and CTO of Within (formerly Klarity), who spent nearly a decade and several pivots searching for product market fit before turning that tool into a company brain that maps how work actually gets done. He breaks down why surviving ten years comes down almost entirely to mindset, why early teams need missionaries instead of people chasing growth metrics, how a simple video-to-documentation tool exposed one of the biggest problems in enterprise, and why product market fit now has to evolve faster than ever. Then he explains why a CEO's own AI fluency sets the ceiling for the whole company, and why moving people from anxiety to abundance matters more than any tool rollout. 🔔 Subscribe for weekly conversations with the founders, CEOs, and VCs rewriting the rules of technology. Chapters: 0:00 - Intro 0:53 - The Long Road To Product Market Fit 2:53 - Why Survival Is All Mindset 4:37 - The Team That Stayed 6:44 - Hiring Missionaries, Not Growth Chasers 9:02 - The Internal Tool Customers Wanted To Buy 13:11 - Building A Company Brain 15:05 - Product Market Fit Is A Moving Target 17:35 - The Power And Risk Of Dogfooding 18:40 - Advice For AI-First CEOs 20:19 - From AI Anxiety To Abundance 22:12 - Book Recommendation Book Recommendations: Americana by Bhu Srinivasan The Happiness Hypothesis by Jonathan Haidt About Nischal Nadhamuni Nischal Nadhamuni is co-founder and CTO of Within, the San Francisco enterprise AI company formerly known as Klarity, which builds a company brain that maps how work actually gets done so organizations can decide where people and AI agents fit. He founded the company in 2017 with Andrew Antos, whom he met at MIT, and went through Y Combinator's Summer 2018 batch. The company spent its early years automating document review, first for legal teams and then for finance and accounting, before an internal process documentation tool became the core of its current platform. Customers include OpenAI, DoorDash, ServiceNow, Salesforce, JLL and McKesson. The company raised an eighteen million dollar Series A led by Tola Capital, followed by a seventy million dollar Series B in 2024 led by Nat Friedman and Daniel Gross, with Scale Venture Partners, Tola Capital, Picus Capital, Invus and Y Combinator participating. Before founding the company, he worked on AI applications at Airware, Flipkart and Massachusetts General Hospital, where his team's machine learning work on detecting cancer-causing genes was published in the Journal of Clinical Oncology. He holds a computer science degree from MIT and was named to Forbes 30 Under 30 in AI. He argues that the hardest part of enterprise AI isn't the model, it's understanding how people actually work. Follow Nischal Nadhamuni on LinkedIn: https://www.linkedin.com/in/nishchaln/  About the show Inside the Silicon Mind takes you behind the scenes with the founders, CEOs, and VCs building the future of technology. Every week, Firas Sozan sits down with the operators and investors rewriting the rules - from deep‑tech startups to the security challenges created by AI. Hosted by Firas Sozan. 🔔 Subscribe for more deep dives on:  - AI and the future of software  - Venture capital and company building  - Developer tools, infrastructure, and engineering careers YouTube: https://www.youtube.com/@InsideTheSiliconMind Apple Podcasts: https://bit.ly/apple-itsm Spotify: https://bit.ly/spotify-itsm Follow the host on LinkedIn: https://www.linkedin.com/in/firassozan/  Website links: https://www.harrisonclarke.com/ https://www.harrisonclarkeventures.com/ https://thepmfplaybook.com/

Eighty-five percent of the world economy runs on physical things, and almost none of today's AI was built to understand them. I sat down with Ivan Poupyrev, co-founder and CEO of Archetype AI, who spent nearly a decade at Google shipping radar and sensor products, worked at Disney and Sony before that, and left to build Newton, a single foundation model for the physical world. He pulls back the curtain on why the next wave of AI will run on sensor data, not language. He breaks down why a car or a turbine can't tell you what's wrong in words and only speaks in measurements, why the fastest wins won't come from robots but from the computers already sitting inside every crane, car and power line, why one general model can replace the patchwork of single-use tools industry pays for today, and why big tech stays late to new markets until they're earning a hundred million dollars a day. Then he gets into why AI raises the stakes on every hire instead of lowering them. 🔔 Subscribe for weekly conversations with the founders, CEOs, and VCs rewriting the rules of technology. Chapters: 0:00 - Intro 1:34 - The Next Trillion Dollar Company 2:28 - What Physical AI Actually Means 4:11 - The Language Of Sensors 5:41 - Where Physical AI Hits First 8:53 - One Model For The Physical World 11:08 - How To Know If You're Too Early 14:12 - The Shrinking Physical Workforce 15:52 - Why Startups Beat Big Tech 19:29 - Building The Founding Team 24:09 - Why AI Makes Every Hire Critical 25:44 - Book Recommendation Book Recommendation: What Is Real? by Adam Becker About Ivan Poupyrev Ivan Poupyrev is co-founder and CEO of Archetype AI, the Palo Alto physical AI company behind Newton, a foundation model that reads real-time sensor data and turns it into insight and action for industries like manufacturing, energy, logistics and transportation. Before founding Archetype AI in 2023 with four of his former Google colleagues, he spent about nine years at Google's Advanced Technology and Projects group, where his team built radar and other complex sensors and shipped them in consumer products, including Project Soli and Project Jacquard. Earlier in his career he worked at Disney Research and Sony. Archetype AI raised a thirteen million dollar seed round led by Venrock, followed by a thirty-five million dollar Series A led by IAG Capital Partners and Hitachi Ventures, with Bezos Expeditions, the Amazon Industrial Innovation Fund and Samsung participating. He argues that the real economy lives in the physical world, and that the AI that learns to understand it will matter more than the AI that understands words. Follow Ivan Poupyrev on LinkedIn: https://www.linkedin.com/in/ivan-poupyrev/ About the show Inside the Silicon Mind takes you behind the scenes with the founders, CEOs, and VCs building the future of technology. Every week, Firas Sozan sits down with the operators and investors rewriting the rules - from deep‑tech startups to the security challenges created by AI. Hosted by Firas Sozan. 🔔 Subscribe for more deep dives on:  - AI and the future of software  - Venture capital and company building  - Developer tools, infrastructure, and engineering careers YouTube: https://www.youtube.com/@InsideTheSiliconMind Apple Podcasts: https://bit.ly/apple-itsm Spotify: https://bit.ly/spotify-itsm Follow the host on LinkedIn: https://www.linkedin.com/in/firassozan/  Website links: https://www.harrisonclarke.com/ https://www.harrisonclarkeventures.com/ https://thepmfplaybook.com

The hardest part of starting over isn't learning something new. It's unlearning everything that made you successful. I sat down with Julia Seregina, an engineer turned tech and industrial executive who ran teams from London to Singapore to Abu Dhabi, then walked away from the big salary to become a first-time founder after thirty in a field where she had no authority at all. Now building in stealth at the edge of AI and science, she pulls back the curtain on what it actually takes to reinvent yourself once you're already an expert. She breaks down why no one has managed to recreate Silicon Valley, why she has to melt old thought patterns before she can think like a founder again, why macro thinking is becoming a more valuable skill than shipping fast, and the three-month test she recommends before anyone quits a career to start a new one. Then she gets into the one thing that has stayed true across every industry and country she has built in. 🔔 Subscribe for weekly conversations with the founders, CEOs, and VCs rewriting the rules of technology. Chapters: 0:00 - Intro 1:18 - Why No One Has Recreated Silicon Valley 4:44 - Why Other Places Punish Failure 6:57 - What You Lose When You Leave The Bay 9:21 - Speed Versus Thinking 10:48 - Founder Challenges AI Created 12:06 - Building AI For Science In Stealth 13:28 - Starting Over As A Newcomer 16:05 - Unlearning Old Patterns 17:22 - Never Too Late, Just Harder 17:59 - What Stays The Same Across Industries 19:22 - What Reinvention Taught Her 20:40 - The First Step If You Feel Trapped 24:04 - Book Recommendation Book Recommendation: A Brief History of Intelligence by Max Bennett About Julia Seregina Julia Seregina is a founder building in stealth in AI for science, working on tools that help biology and life sciences researchers do better work and on AI that reasons like a scientist over long time horizons rather than simply automating steps. She began her career as an engineer and rose quickly into executive roles at technology and industrial companies, working across London, Italy, Poland, Central Asia, Singapore, Southeast Asia, Abu Dhabi and Berlin. She then left corporate life to found companies in AI for healthcare and life sciences, a field she had no formal training in. She has been part of the South Park Commons community and splits her time between the Bay Area and Abu Dhabi. She argues that reinvention starts with unlearning, that conviction and great talent matter more than chasing whatever trend is hot, and that the smartest way to test a new chapter is to build on the side, time-box it, and see whether your conviction grows with every conversation. Follow Julia Seregina on LinkedIn: https://www.linkedin.com/in/julia-seregina-4a7a30142/ About the show Inside the Silicon Mind takes you behind the scenes with the founders, CEOs, and VCs building the future of technology. Every week, Firas Sozan sits down with the operators and investors rewriting the rules - from deep‑tech startups to the security challenges created by AI. Hosted by Firas Sozan. 🔔 Subscribe for more deep dives on:  - AI and the future of software  - Venture capital and company building  - Developer tools, infrastructure, and engineering careers YouTube: https://www.youtube.com/@InsideTheSiliconMind Apple Podcasts: https://bit.ly/apple-itsm Spotify: https://bit.ly/spotify-itsm Follow the host on LinkedIn: https://www.linkedin.com/in/firassozan/ Website links: https://www.harrisonclarke.com/ https://www.harrisonclarkeventures.com/ https://thepmfplaybook.com/

The thing that got you your first twenty customers is the exact thing that will break your company at two hundred. I sat down with the founder and CEO of Concentric AI, who started the company three months before the world shut down, raised a forty five million dollar Series B and acquired two companies at once to keep pace with AI, and he pulls back the curtain on what it actually costs a founder to become a CEO. He breaks down the three versions of himself he has had to become, the stress versus frustration framework a VC gave him that now decides which environments suit which people, why context switching quietly destroys a founder's judgment, and why deliberate hiring beats headcount because three hires never give you three people's productivity. Then he gets into the part that keeps him up at night, staying relevant to a market that can wipe out an entire category overnight. 🔔 Subscribe for weekly conversations with the founders, CEOs, and VCs rewriting the rules of technology. Chapters: 0:00 - Intro 0:46 - Advice At One Million ARR 4:30 - The Cost Of Context Switching 7:59 - Becoming Three Different CEOs 13:16 - What Worked At Zero To One 15:42 - When A Founder Becomes A CEO 18:40 - Stress Versus Frustration 22:48 - Hiring Without Adding Frustration 24:49 - Keeping Your Finger On PMF 28:00 - The Problem That Never Goes Away 32:06 - Culture Is Doing Not Saying 35:56 - Book Recommendations Book Recommendations: Thinking, Fast and Slow by Daniel Kahneman Team of Rivals by Doris Kearns Goodwin About Karthik Krishnan Karthik Krishnan is founder and CEO of Concentric AI, the data security platform that uses AI to find, classify and protect sensitive data across cloud and on premises environments. He spent decades in product before starting the company, including roles at Intel, Juniper Networks, PGP Corporation and Embrane, and was part of the founding team at security analytics company Niara, which was acquired by HPE, where he then served as vice president of product management. At Concentric AI he has raised a forty five million dollar Series B led by Top Tier Capital Partners and HarbourVest Partners, acquired Swift Security and Acante in a single move to bring GenAI governance and AI data curation into the platform, and built the company into a business selling directly to enterprise CISOs. He is one of the clearest voices on the founder to CEO transition, arguing that the job is to stay intellectually honest about whether you are still the right person to lead, to protect your company from frustration for as long as possible, and to keep your head outside the building where the market actually decides your fate. Follow Karthik Krishnan on LinkedIn: https://www.linkedin.com/in/kkrishnan/ About the show Inside the Silicon Mind takes you behind the scenes with the founders, CEOs, and VCs building the future of technology. Every week, Firas Sozan sits down with the operators and investors rewriting the rules - from deep‑tech startups to the security challenges created by AI. Hosted by Firas Sozan. 🔔 Subscribe for more deep dives on:  - AI and the future of software  - Venture capital and company building  - Developer tools, infrastructure, and engineering careers YouTube: https://www.youtube.com/@InsideTheSiliconMind Apple Podcasts: https://bit.ly/apple-itsm Spotify: https://bit.ly/spotify-itsm Follow the host on LinkedIn: https://www.linkedin.com/in/firassozan/ Website links: https://www.harrisonclarke.com/ https://www.harrisonclarkeventures.com/ https://thepmfplaybook.com

What if the SaaS apocalypse everyone spent two years predicting was just a discounted entry point? I sat down with the CEO of Clari + Salesloft, a thousand person company built from the merger of two category leaders, now serving over 5,000 customers on roughly $450 million in combined ARR, and he pulls back the curtain on why the AI native land grab is already unravelling. He explains why he gets pitched at least one struggling AI go to market startup every single day, why adoption and retention stall them long before the next funding round, and why the survivors will be the ones that own an entire workflow rather than a single task. He also dismantles the buy versus build story, exposes why most acquisitions quietly kill the culture they paid for, and lays out the three things he tracks to run a business at this scale. 🔔 Subscribe for weekly conversations with the founders, CEOs, and VCs rewriting the rules of technology. Chapters: 0:00 - Intro 1:06 - Is SaaS Actually Dead 3:44 - Signs Buyers Now Want Real ROI 5:33 - What Separates AI Winners From Losers 6:21 - Will AI Replace Humans 8:00 - The Buy Versus Build Trap 10:16 - The Coming AI Consolidation Wave 12:52 - Why Most Acquisitions Fail 15:53 - Advice For Founders Planning An Exit 17:29 - Bringing Clari And Salesloft Together 21:18 - Will It Make The Boat Go Faster 22:57 - The Three Things He Tracks Book Recommendations: Will It Make The Boat Go Faster? by Ben Hunt-Davis and Harriet Beveridge About Steve Cox: Steve Cox is Chief Executive Officer of Clari + Salesloft, the Revenue AI company formed when Clari and Salesloft completed their merger on 3 December 2025, where he was appointed CEO and joined two weeks after close. He brings more than 25 years of SaaS leadership with deep expertise in scaling businesses and M&A driven transformation, most recently as CEO of Employ, parent company of JazzHR, Lever, Jobvite and NXTThing RPO, and before that as COO of Community Brands and in executive roles at Vertafore, Chess and TSG. He started his career as a hands on technologist before moving into operations and leadership, and now oversees a combined organisation of roughly a thousand people, a doubled R&D investment, and a platform ingesting more than 10 billion revenue interactions and 1 trillion data signals to power what the company calls the Predictive Revenue System. He is one of the clearest voices on what actually happens after the AI hype curve, arguing that point solutions cannot survive without end to end workflow coverage, that historical data is the only real moat for model accuracy, and that in a few years nobody will say AI company at all because they will just be software companies again. Follow Steve Cox on LinkedIn: https://www.linkedin.com/in/steve-cox-588a2024/ About the show Inside the Silicon Mind takes you behind the scenes with the founders, CEOs, and VCs building the future of technology. Every week, Firas Sozan sits down with the operators and investors rewriting the rules - from deep‑tech startups to the security challenges created by AI. Hosted by Firas Sozan. 🔔 Subscribe for more deep dives on:  - AI and the future of software  - Venture capital and company building  - Developer tools, infrastructure, and engineering careers YouTube: https://www.youtube.com/@InsideTheSiliconMind Apple Podcasts: https://bit.ly/apple-itsm Spotify: https://bit.ly/spotify-itsm Follow the host on LinkedIn: https://www.linkedin.com/in/firassozan/ Website links: https://www.harrisonclarke.com/ https://www.harrisonclarkeventures.com/ https://thepmfplaybook.com/

If your product is a wrapper on top of an LLM, is your company already over? I sat down with the cofounder and CTO of Rox, the revenue agent platform backed by Sequoia, General Catalyst and Google Ventures and used by teams at OpenAI, Ramp, Confluent, MongoDB and NVIDIA, and he does not hedge on where any of this is heading. He breaks down the death of the traditional engineering org and the rise of the product engineer, why Rox runs with zero product managers, why he hires for slope rather than intercept, and how a knowledge graph built across CRM, support and product usage data turns messy enterprise systems into agents that run outbound end to end. Then he calls the market as he sees it, a barbell where companies without a real moat are facing serious valuation compression while the big labs and the deep infrastructure players take the spoils. Subscribe for weekly conversations with the founders, CEOs, and VCs rewriting the rules of technology. Chapters: 0:00 - Intro 0:42 - Rebuilding Rox With Ten Engineers 2:37 - Why Rox Has No Product Managers 4:04 - Slope Versus Intercept 9:40 - Skills AI Has Not Replaced 11:49 - Why Revenue Agents Beat Coding Agents 13:50 - The Problem Rox Solves 16:22 - Building The Knowledge Graph 18:29 - Running Rox On Rox 20:33 - What Founders Should Build Next 23:20 - Are We In An AI Bubble 28:58 - Book Recommendation Book Recommendations: Extreme Ownership by Jocko Willink and Leif Babin About Shriram Sridharan Shriram Sridharan is cofounder and CTO of Rox, the AI revenue agent platform built for Fortune 2000 revenue teams and backed by Sequoia Capital, General Catalyst and Google Ventures. He was a founding engineer on Amazon Aurora and an early engineer at Confluent, where he built systems behind a business that scaled from zero to billions in revenue, before starting Rox with CEO Ishan Mukherjee. At Rox he owns the platform layer, the knowledge graph that pulls CRM, support, product usage, transcripts and email into a single place, and the turnkey agents that sit on top of it to automate inbound, outbound, deal management and renewals for customers including OpenAI, Ramp, Confluent, MongoDB and NVIDIA. He is one of the sharpest voices on what AI actually does to an engineering organisation, arguing that product management is collapsing into the engineer, that hiring should reward slope over intercept, and that the last skills to be automated are taste, judgment and the ability to trust what an agent just shipped. Follow Shriram Sridharan on LinkedIn: https://www.linkedin.com/in/shriramsridharan/ About the show Inside the Silicon Mind takes you behind the scenes with the founders, CEOs, and VCs building the future of technology. Every week, Firas Sozan sits down with the operators and investors rewriting the rules - from deep‑tech startups to the security challenges created by AI. Hosted by Firas Sozan. Subscribe for more deep dives on: AI and the future of software Venture capital and company building Developer tools, infrastructure, and engineering careers YouTube: https://www.youtube.com/@InsideTheSiliconMind  Apple Podcasts: https://bit.ly/apple-itsm  Spotify: https://bit.ly/spotify-itsm  Follow the host on LinkedIn: https://www.linkedin.com/in/firassozan/  Website links: https://www.harrisonclarke.com/  https://www.harrisonclarkeventures.com/  https://thepmfplaybook.com/

Why do the best investors almost always back the business operator over the brilliant technologist? I sat down with a venture investor who has been deploying capital since the early days of the internet and has invested through the internet, mobile and cloud waves and now AI, and he pulls back the curtain on how he separates a real technology shift from an overheated segment. This conversation dismantles the idea that AI is just another hype cycle and exposes where the money is actually moving next. 🔔 Subscribe for weekly conversations with the founders, CEOs, and VCs rewriting the rules of technology. Chapters: 0:00 - Intro 0:43 - Revolution or Speculative Bubble 5:19 - Comparing AI to Cloud 8:31 - Why Hardware Matters Again 10:48 - From Speculation to Conviction 17:43 - The Messy Middle 18:51 - Why Private Money Floods AI 21:12 - What a Curve Shift Means 25:37 - Betting on Founders Not Tech 31:16 - Teams That Work Like Bands 36:00 - The Rise of Citizen Engineers 40:42 - Book Recommendation Book Recommendation: The Three-Body Problem by Liu Cixin About Mark Rostick Mark Rostick is Vice President and Senior Managing Director at Intel Capital, where he has been investing since 1999 and serves as a voting member of the investment committee. He helps manage the firm's cloud portfolio and has spent more than two decades backing cloud applications, infrastructure hardware and software, and AI, with holdings including MinIO, Immuta, Lilt, Opaque Systems, Cherre, Tetrate and Beep, and exits including Virtustream, Velostrata, Kazan Networks, Nyansa and Amplidata. He came to venture from careers in law and banking, holds an MBA and a JD from the University of South Carolina, and has sat as a director or board observer across most of the companies he has funded. He is one of the sharpest voices on the economics of technology waves, arguing that the real money is made when the entire curve shifts rather than when capital simply moves along it, and that in enterprise AI the fortunes will be won in the messy middle where adoption, not invention, is the hard part. Follow Mark Rostick on LinkedIn: https://www.linkedin.com/in/mrostick/ About The Show Inside the Silicon Mind takes you behind the scenes with the founders, CEOs, and VCs building the future of technology. Every week, Firas Sozan sits down with the operators and investors rewriting the rules - from deep‑tech startups to the security challenges created by AI. Hosted by Firas Sozan. Powered by Harrison Clarke. 🔔 Subscribe for more deep dives on:  - AI and the future of software  - Venture capital and company building  - Developer tools, infrastructure, and engineering careers YouTube: https://www.youtube.com/@InsideTheSiliconMind Apple Podcasts: https://bit.ly/apple-itsm Spotify: https://bit.ly/spotify-itsm Follow the host on LinkedIn: https://www.linkedin.com/in/firassozan/ Website links: https://www.harrisonclarke.com/ https://www.harrisonclarkeventures.com/ https://thepmfplaybook.com/

Why does Nvidia stock drop every time a better open source model ships? Benny Chen thinks the market has this exactly backwards, and he has the track record to argue it. He spent years on ASICs and GPU model serving inside Meta before co-founding Fireworks AI in September 2022, raising from Benchmark almost immediately, and going on to help Cursor tune Composer 2 and Composer 2.5, help Cognition tune their model, and work with Harvey on a legal model. In this conversation we dismantle the idea that open source progress destroys GPU demand, and pull back the curtain on what actually breaks if this cycle unwinds. He breaks down why collaboration is always cheaper than closed research, why the recent DeepSeek release undercuts everything in its quality class on price, and how the models his team serves scaled a thousand times in four years. We get into the exact customization playbook Fireworks runs, taking an open source base model, tuning it to price competitiveness against frontier models, then building a data flywheel so a business can keep improving that model and swap the base out as better releases land. He also exposes why safety filters are asymmetrical, using the Hugging Face incident to show how attackers route around prompt guardrails while defenders cannot, and makes the case that this looks less like the dot com bubble and more like a real estate bubble, where the demand is real and leverage is the thing that takes people down. 🔔 Subscribe for weekly conversations with the founders, CEOs, and VCs rewriting the rules of technology. Chapters:  - 0:00 - Intro  - 1:22 - Why Open Source Wins  - 3:43 - Why Nvidia Drops Make No Sense  - 4:30 - Founding Fireworks  - 6:37 - The First Wave of GPU Demand  - 8:51 - The Founding Team  - 10:48 - Betting on Open Source Early  - 13:26 - What Sam Altman Would Disagree With  - 15:48 - Control, Filters and Access  - 19:08 - Is AI a Bubble  - 22:11 - Depending on the Frontier Labs  - 24:52 - Fireworks Product Market Fit  - 27:42 - Book Recommendation Book Recommendation: Principles Series by Ray Dalio About Benny Chen Benny Chen is a co-founder of Fireworks AI, the inference and post training platform that helps companies customize open source models, serve them in production, and own the intelligence their products run on instead of renting it from a frontier lab. He spent years working on ASIC and GPU model serving inside Meta, where he was an ads infrastructure lead, before cofounding Fireworks in September 2022 with a team drawn largely from PyTorch and PyTorch adjacent orgs and backing from Benchmark. His team has tuned models with Cursor for Composer 2 and Composer 2.5, and worked with Cognition, Harvey, Vercel, GenSpark, Factory AI, and cybersecurity firms deploying models on their own infrastructure. He is one of the clearest voices arguing that better open source models make GPUs more valuable rather than less, and that every serious business should be building its own data flywheel rather than depending on a single provider for access. Follow Benny Chen on LinkedIn: https://www.linkedin.com/in/benny-yufei-chen-2238575a/ About The Show Inside the Silicon Mind takes you behind the scenes with the founders, CEOs, and VCs building the future of technology. Every week, Firas Sozan sits down with the operators and investors rewriting the rules - from deep‑tech startups to the security challenges created by AI. Hosted by Firas Sozan. Powered by Harrison Clarke. 🔔 Subscribe for more deep dives on:  - AI and the future of software  - Venture capital and company building  - Developer tools, infrastructure, and engineering careers YouTube: https://www.youtube.com/@InsideTheSiliconMind      Apple Podcasts: https://bit.ly/apple-itsm  Spotify: https://bit.ly/spotify-itsm   Follow the host on LinkedIn:  https://www.linkedin.com/in/firassozan/  Website links:  https://www.harrisonclarke.com/  https://www.harrisonclarkeventures.com/  https://thepmfplaybook.com/

Everyone keeps repeating that AI is making cyber defenders faster and more productive, but in this episode my guest Evan Powell argues the opposite is quietly true, that AI is handing the real advantage to the attackers. He has spent decades building deep tech companies and now runs Deep Tempo, where his team built a purpose trained LogLM that reads nothing but logs and flags advanced attacks with very low false positives. Across our conversation he dismantles the theater of cyber, the idea that hundreds of billions in annual security spending is actually buying us safety, and he opens up about the deepfake that came through his own inbound before compliance caught it. You will hear why the signature based model the industry was built on now runs on what he calls negative time, how attackers armed with distilled models hunt novel vulnerabilities before anyone has built a detection, and why bolting a reasoning model onto defense produces high cost and poor accuracy at real world volume. He exposes the innovator's dilemma that pushes vendors toward steak dinners and F1 sponsorships instead of system level thinking, makes the case for open source tools like his AI SOC project Vigil, and lays out the risk every founder building on a single LLM faces if that model gets switched off. He also gets candid about running ninety hour weeks, tying his focus to one customer outcome at a time, and why he believes reading matters more now than ever. 🔔 Subscribe for weekly conversations with the founders, CEOs, and VCs rewriting the rules of technology. Chapters: 0:00 - Intro 1:00 - The Theater of Cyber Spending 2:07 - A Deepfake Hit Our Business 3:04 - Why Defenders Have Negative Time 5:01 - Cyber Wasn't Ready for This 8:38 - What RSA Vendors Get Wrong 10:49 - Already Losing Before AI 12:01 - Why LLMs Can't Detect Attacks 16:46 - Inside DeepTempo 20:05 - Building in a Golden Age 21:52 - What If the Models Switch Off 25:46 - Book Recommendation  29:38 - Wrap Up Book Recommendations: The Structure of Scientific Revolutions by Thomas Kuhn The Innovator's Dilemma by Clayton Christensen About Evan Powell Evan Powell is the CEO and founder of DeepTempo, the company building an AI native detection intelligence layer for cybersecurity powered by LogLM, a foundation model pre trained purely on logs to identify attacker intent without relying on signatures. A serial deep tech entrepreneur who previously built StackStorm and other Linux Foundation and CNCF projects, he also launched the open source AI SOC Vigil and the open source SOCBench benchmark. His approach has been proven with reference partners spanning telco, banking, national government, and critical infrastructure, and he is now bringing the technology to MSSPs and thousands of practitioners through a system level, uncomfortably transparent, outcome tied model. Follow Evan Powell on LinkedIn: https://www.linkedin.com/in/epowell About The Show Inside the Silicon Mind takes you behind the scenes with the founders, CEOs, and VCs building the future of technology. Every week, Firas Sozan sits down with the operators and investors rewriting the rules - from deep‑tech startups to the security challenges created by AI. Hosted by Firas Sozan. Powered by Harrison Clarke. 🔔 Subscribe for more deep dives on: AI and the future of software Venture capital and company building Developer tools, infrastructure, and engineering careers YouTube: https://www.youtube.com/@InsideTheSiliconMind  Apple Podcasts: https://bit.ly/apple-itsm  Spotify: https://bit.ly/spotify-itsm  Follow the host on LinkedIn: https://www.linkedin.com/in/firassozan/ Website links:  https://www.harrisonclarke.com/  https://www.harrisonclarkeventures.com/  https://thepmfplaybook.com/

What if the worst moment of your life, the one you are most ashamed of, turned out to be the exact thing that made everything after it possible? In this episode of Inside the Silicon Mind, Jared Shepard pulls back the curtain on a journey that runs from a homeless sixteen year old sleeping in the back of a baby blue Lincoln behind a Sinclair gas station to the founder and CEO seat of Hypori, a company he built from a failing business he bought for one million dollars into a venture worth over three hundred million. He dismantles the myth that the American dream is dead, tracing how the Army rebuilt his moral compass, how a judge who gave him thirty days to produce an enlistment contract changed his life, and how he founded Intelligent Waves in 2006 on a spreadsheet in Baghdad.  You will learn how he spotted the twelve patents inside a company being liquidated, hired six of its original developers, and turned a denied space communications problem into a zero trust, zero data at rest platform that pulls four gigs of bandwidth over a phone by streaming a full operating system from the cloud for under fifteen dollars a month. Jared also breaks down why passion is the one thing AI can never replicate, why money is a side effect and not the goal, and how becoming a sponge for mentorship and owning every outcome built the discipline behind everything he has done. 🔔 Subscribe for weekly conversations with the founders, CEOs, and VCs rewriting the rules of technology. Chapters: 0:00 - Intro 1:00 - Is the American Dream Dead 2:09 - From Homeless Teenager 7:47 - The Values That Set Him Apart 8:43 - The Choice to Join the Army 12:49 - Starting Intelligent Waves in Baghdad 14:45 - Buying a Failing Company 16:18 - The Privacy Epiphany 20:12 - The Billion Dollar Vision 24:06 - Advice to Founders 29:37 - Book Recommendations 32:08 - Wrap Up Book recommendation: Founder vs Investor for anyone weighing whether to take outside money, and Build by Tony Fadell for how to lead, focus a team, and create something unique.  About Jared Shepard: Jared Shepard is the President and CEO of Hypori, the virtual mobility company delivering secure edge access through a zero trust, zero data at rest architecture, and the largest BYOD vendor in the Department of War. A U.S. Army veteran who served as Lead Technical Planner for III Corps and deployed across Iraq and Afghanistan, he founded Intelligent Waves in 2006 to support the Joint IED Defeat Organization and later spun Hypori out as its own company in 2021. He is also the founder and chairman of Warriors Ethos, a veteran focused nonprofit dedicated to the career transition of wounded service members, veterans, and their families. He builds with a conviction that impact comes first and money follows, carrying the lessons of his hardest years into every company he leads. Follow Jared Shepard on LinkedIn:  https://www.linkedin.com/in/shepardj  About the show Inside the Silicon Mind takes you behind the scenes with the founders, CEOs, and VCs building the future of technology. Every week, Firas Sozan sits down with the operators and investors rewriting the rules - from deep‑tech startups to the security challenges created by AI. Hosted by Firas Sozan. Powered by Harrison Clarke. 🔔 Subscribe for more deep dives on: • AI and the future of software • Venture capital and company building • Developer tools, infrastructure, and engineering careers YouTube: https://www.youtube.com/@InsideTheSiliconMind    Apple Podcasts: https://bit.ly/apple-itsm  Spotify: https://bit.ly/spotify-itsm  Follow the host on LinkedIn:  https://www.linkedin.com/in/firassozan/ Website links:  https://www.harrisonclarke.com/  https://www.harrisonclarkeventures.com/  https://thepmfplaybook.com/

What if a clinical record sells for magnitudes more on the dark web than a social security number, and almost nobody outside healthcare is talking about it? In this episode of Inside the Silicon Mind, Sandeep pulls back the curtain on building Heald, a company attacking metabolic health as the root cause behind most chronic conditions, drawing on a career that took him from Amazon to founder and investor. He dismantles the founder myth that speed always wins, making the case that patience means slowing down enough to decide on data, research, and customer signals rather than gut feel, then letting the decision stick long enough to read the KPIs. You will learn where impulse still earns its place in hiring and culture, why failing to fire fast is a disservice to everyone who took a risk on your company, and how AI belongs inside a healthcare product only where there is a real problem for it to solve.  🔔 Subscribe for weekly conversations with the founders, CEOs, and VCs rewriting the rules of technology. Chapters: 0:00 - Intro 0:58 - Why Patience Beats Speed 2:53 - Balancing Patience With Momentum 5:03 - Decisions You Regret Most 7:26 - Turning Regret Into Drive 8:07 - When to Trust Your Impulse 9:54 - Hire Slow Fire Fast 12:07 - Letting People Self Select Out 13:07 - Does AI Terrify or Excite You 14:37 - Protecting Healthcare Data 16:03 - What Heald Actually Does 19:15 - Metabolic Health Explained 20:42 - Sleep HRV and Blood Glucose 23:28 - Intermittent Fasting Is Personal 24:36 - Book Recommendation Book Recommendation: Shoe Dog by Phil Knight About Sandeep Sandeep is the founder of Heald, a company built to address metabolic health, the root cause behind most chronic conditions from obesity and hypertension to cardiac disease and cancer. Heald combines technology with human care teams across four factors most of the industry ignores: nutrition, fitness, sleep, and stress, pulling data from wearables, glucose monitors, and smart scales while keeping humans accountable to patients. Before founding Heald he spent time at Amazon, where the leadership principle that leaders are right a lot shaped how he balances instinct with evidence. He builds with a bias toward data over impulse, patience over speed, and using AI where there is an actual problem to solve rather than as a solution in search of one. LinkedIn: https://www.linkedin.com/in/misrasandeep/ About The Show Inside the Silicon Mind takes you behind the scenes with the founders, CEOs, and VCs building the future of technology. Every week, Firas Sozan sits down with the operators and investors rewriting the rules, from deep tech startups to the security challenges created by AI. Hosted by Firas Sozan. Powered by Harrison Clarke. 🔔 Subscribe for more deep dives on: AI and the future of software Venture capital and company building Developer tools, infrastructure, and engineering careers YouTube: https://www.youtube.com/@InsideTheSiliconMind Apple Podcasts: https://bit.ly/apple-itsm Spotify: https://bit.ly/spotify-itsm Follow the host on LinkedIn: https://www.linkedin.com/in/firassozan/ Website links: https://www.harrisonclarke.com/ https://www.harrisonclarkeventures.com/ https://thepmfplaybook.com/

What happens to observability when the person staring at the dashboard can no longer understand the code an AI just wrote? Anish Agarwal, CEO and co-founder of Traversal and a Columbia professor with an MIT PhD, spent two and a half years building an agentic AI SRE platform for enterprises that generate roughly a petabyte of telemetry a day, and in this conversation he dismantles the twenty year old assumption that a human is watching the dashboard and exposes why that model collapses at industrial scale. You will see why dashboards are really just enshrined queries built to explain the past instead of predicting novel failures, and why causal AI matters when one broken service spreads like an epidemic and every alert becomes a fake needle in a haystack   🔔 Subscribe for weekly conversations with the founders, CEOs, and VCs rewriting the rules of technology.   Chapters: 0:00 - Intro 1:18 - Why Dashboards Are Breaking 3:47 - Building Traversal AI SRE 4:37 - Research Roots in Causal AI 7:29 - Why Startups Beat Academia Now 8:44 - What Happens to SREs 11:12 - The End of the War Room 13:16 - Are Dashboards Obsolete 14:47 - The Rise of the SR Researcher 17:35 - Why Incumbents Cannot Win 20:16 - Talent Density as the Moat 25:18 - Book Recommendation   Book Recommendation: Mostly Harmless Econometrics   About Anish Agarwal: Anish is the CEO and co-founder of Traversal, an agentic AI SRE platform that helps enterprises troubleshoot and self-heal complex systems at a scale where telemetry can reach a petabyte of data a day. A professor at Columbia University with a PhD from MIT, he built the company alongside co-founders whose research spans causal AI and reinforcement learning, translating that academic depth into a product that rethinks observability from the ground up for a world where AI writes the code. He is passionate about removing the most painful parts of the site reliability engineer's job so that engineers can get back to architecting resilient systems rather than firefighting incidents.   Follow Anish Agarwal on LinkedIn: https://www.linkedin.com/in/anish-agarwal-io/   About the show Inside the Silicon Mind takes you behind the scenes with the founders, CEOs, and VCs building the future of technology. Every week, Firas Sozan sits down with the operators and investors rewriting the rules - from deep‑tech startups to the security challenges created by AI. Hosted by Firas Sozan. Powered by Harrison Clarke.   🔔 Subscribe for more deep dives on: AI and the future of software Venture capital and company building Developer tools, infrastructure, and engineering careers   YouTube: https://www.youtube.com/@InsideTheSiliconMind    Apple Podcasts: https://bit.ly/apple-itsm Spotify: https://bit.ly/spotify-itsm    Follow the host on LinkedIn:  https://www.linkedin.com/in/firassozan/   Website links:  https://www.harrisonclarke.com/  https://www.harrisonclarkeventures.com/  https://thepmfplaybook.com/    #AI #InsideTheSiliconMind #Observability #SRE #CausalAI

What if the terms everyone is obsessing over in AI are the exact reason your startup is missing the big picture? In this episode of Inside the Silicon Mind, Farshid Sabet pulls back the curtain on building AI companies the right way, drawing on a career spanning SanDisk, Micron, Movidius, and Aptina before founding Corvic in the B2B enterprise space. He dismantles the hype cycles that have distracted founders since 2022, from building your own models to chasing specific GPUs to fixating on prompting, RAG, context, and now agentics. You will learn how to tell a genuine productivity gain from noise, why autonomous agents matter without deserving every buzzword thrown at them, and how to keep your eyes on business fundamentals while the market chases trends.   🔔 Subscribe for weekly conversations with the founders, CEOs, and VCs rewriting the rules of technology.   Chapters: 0:00 - Intro 1:10 - The Biggest Learning Moment 2:07 - Hype Cycles in AI 3:07 - The Agentic AI Shift 4:50 - The Real Big Picture 6:17 - The Hardest Nos to Make 9:58 - Where Discipline Comes From 11:51 - Hiring the Right Team 14:19 - The Biggest Scaling Mistake 15:51 - A Book Every Founder Needs 17:39 - Wrap Up   Book recommendation: The Hard Thing About Hard Things by Ben Horowitz   About Farshid Sabet: Farshid Sabet is the founder of Corvic, an AI company focused on the B2B enterprise space, where he helps large organizations across healthcare, banking, retail, and industrial sectors put AI to work. His career spans hardware, software, and AI at companies including SanDisk, Micron, Aptina, and Movidius, where he worked alongside industry leaders like Sanjay Mehrotra and Dr. Eli Harari. He builds with a bias toward business fundamentals over hype, prioritizing scalable products and the discipline to say no when short term wins threaten long term goals.   Follow Farshid Sabet on LinkedIn: https://www.linkedin.com/in/farshid-sabet-298526/   Watch the episode on YouTube: https://youtu.be/7lVmRN1rL8I   About the show Inside the Silicon Mind takes you behind the scenes with the founders, CEOs, and VCs building the future of technology. Every week, Firas Sozan sits down with the operators and investors rewriting the rules, from deep-tech startups to the security challenges created by AI. Hosted by Firas Sozan. Powered by Harrison Clarke.   🔔 Subscribe for more deep dives on: AI and the future of software Venture capital and company building Developer tools, infrastructure, and engineering careers YouTube: https://www.youtube.com/@InsideTheSiliconMind Apple Podcasts: https://bit.ly/apple-itsm Spotify: https://bit.ly/spotify-itsm Follow the host on LinkedIn: https://www.linkedin.com/in/firassozan/ Website links: https://www.harrisonclarke.com/ https://www.harrisonclarkeventures.com/ https://thepmfplaybook.com/ #AI #AgenticAI #StartupFounders #EnterpriseAI #InsideTheSiliconMind

Ex-Google Insider Reveals The Future Of AI in 2026…   What happens to a trillion-dollar AI industry when the intelligence it sells becomes something you can own outright, running on your laptop, impossible to throttle, degrade, or switch off? In this episode of Inside the Silicon Mind, Firas Sozan sits down with Jack O’Brien, ex-Googler and founder of Subconscious, to dismantle the assumption that frontier labs will monopolize intelligence, revealing why 95% of economically valuable AI work can already be done by small open source models and why the “PhD in your pocket” arrives years ahead of schedule.   Subscribe for weekly conversations with the founders, CEOs, and VCs rewriting the rules of technology.   Chapters: 0:00 - The PhD in Your Pocket Thesis 1:41 - Passing the Frontier Threshold 3:40 - Replacing the Chatbot Subscription 5:07 - The 95% Rule of Small Models 7:59 - Compressing the Expertise Premium 10:24 - The Qwen Inflection Point 12:49 - Valuation Reality Check 14:23 - The Application Layer Land Grab 17:15 - Ranking the Big Three 19:10 - Amazon as Switzerland, Apple's Endgame 21:25 - Inside Subconscious: Agentic Workloads 25:20 - Latent Memory Compression 26:21 - Closing Thoughts & Book Recommendation   Book recommendation: The Dark Forest, by Cixin Liu   About Jack O’Brien: Jack O’Brien is the founder of Subconscious, an AI company retraining open source language models to run long-horizon agentic workloads at a fraction of frontier cost. A former Googler who spent years on the search team, he brings deep expertise in machine learning infrastructure, model post-training, and efficient inference. At Subconscious, he focuses on compressing frontier-level intelligence down to models small enough to run on laptops and workstations, enabling businesses and individuals to own their own intelligence rather than rent it through subscriptions. His thesis that 95% of economically valuable AI work can already be handled by small open models sits at the center of his work building the owned-intelligence era.   Follow Jack O’Brien on LinkedIn: https://www.linkedin.com/in/thejackobrien/   About the show Inside the Silicon Mind takes you behind the scenes with the founders, CEOs, and VCs building the future of technology. Every week, Firas Sozan sits down with the operators and investors rewriting the rules - from deep-tech startups to the security challenges created by AI. Hosted by Firas Sozan. Powered by Harrison Clarke.   Subscribe for more deep dives on: • AI and the future of software • Venture capital and company building • Developer tools, infrastructure, and engineering careers   YouTube: https://www.youtube.com/@InsideTheSiliconMind Apple Podcasts: https://bit.ly/apple-itsm Spotify: https://bit.ly/spotify-itsm   Follow the host on LinkedIn: https://www.linkedin.com/in/firassozan/   Website links: https://www.harrisonclarke.com/ https://www.harrisonclarkeventures.com/ https://thepmfplaybook.com/   #AI #OpenSource #FutureOfAI #LocalAI #InsideTheSiliconMind

Ex-Apple Insider: No One Knows What's Coming With AI   What skills do you teach your children when the career path you spent your life building is on the verge of being entirely automated? If an eight-year-old can already see AI taking over coding, art, and storytelling, how do we prepare the next generation to thrive, not just survive, in a world where everything can be automated in less than twenty years?   In this episode: Why emotional intelligence (EQ) and feelings are not the permanent human moats we think they are, as frontier models begin developing their own functional emotion centers.   The shift from the social media era to the attention media era, where the ultimate currency is the human relationship and the ability to command attention.   How to introduce children to AI capabilities early within controlled, fun environments to prevent them from becoming overwhelmed or left behind by technology.   Why the most critical skill for the next twenty years is not what you learn, but your fundamental learning rate and your speed of adapting to changes.   Subscribe for weekly conversations with the founders, CEOs, and VCs rewriting the rules of technology.   Chapters: 0:00 - The Illusion of Predicting the AI Future 1:46 - Co-Parenting with AI: Planning Birthdays and Playing D&D with Agents 5:14 - The Doomer vs. Optimist Debate: Will Our Species Thrive? 8:54 - The Myth of the EQ Moat: Replicating Human Emotion with Data 11:04 - The Real-World Bottleneck: Clumsy Robots vs. Digital Automation 12:23 - Life in 20 Years: Total Automation and the Crisis of Identity 14:50 - The Attention Economy: Why We Still Value Human Effort 17:03 - Training the Super-Humans: The Economy of Human Feedback Data 19:16 - The Ultimate Moat: High Adaptation Rates over Fixed Skills 21:34 - Sci-Fi and Startup Survival: Adaptability Lessons from Andy Weir 27:08 - Closing Thoughts & Guest Book Recommendations   Book recommendation: Project Hail Mary, by Andy Weir   About Victor: Victor is an AI-focused Chief Technology Officer (CTO) and engineering leader with extensive experience building advanced software architectures and navigating rapid technological disruptions. Immersed deeply in the frontier developments of machine learning, data engineering, and automation, Victor applies a rigorous, systematic approach to technology both in the enterprise sandbox and at home. As a parent and technical operator, he focuses on how rapid adaptation, tool usage, and systemic problem-solving will redefine human capability over the next two decades.   Follow Victor on LinkedIn: https://www.linkedin.com/in/victoraugusteo/   About the show Inside the Silicon Mind takes you behind the scenes with the founders, CEOs, and VCs building the future of technology. Every week, Firas Sozan sits down with the operators and investors rewriting the rules - from deep-tech startups to the security challenges created by AI. Hosted by Firas Sozan. Powered by Harrison Clarke.   Subscribe for more deep dives on: • AI and the future of software • Venture capital and company building • Developer tools, infrastructure, and engineering careers   YouTube: https://www.youtube.com/@InsideTheSiliconMind Apple Podcasts: https://bit.ly/apple-itsm Spotify: https://bit.ly/spotify-itsm   Follow the host on LinkedIn: https://www.linkedin.com/in/firassozan/   Website links: https://www.harrisonclarke.com/ https://www.harrisonclarkeventures.com/ https://thepmfplaybook.com/   #AI #FutureOfWork #ParentingAndAI #AttentionEconomy #InsideTheSiliconMind