Unicorn Builders

Goodfire is an AI research lab working to understand how neural networks actually work, so models can be intentionally designed instead of trained through trial and error.In a recent episode of BUILDERS, we sat down with Eric Ho, Co-founder and CEO of Goodfire, to learn how an exploration-driven interpretability lab is turning foundational research into an enterprise product, and how one public AI incident changed the conversation around its category.Topics Discussed:Why Eric left the SaaS company he spent seven years building after ChatGPT launchedWhy even the researchers building frontier models can’t explain how they workWhy Goodfire modeled itself on Bell Labs, Xerox PARC, and early OpenAI instead of a SaaS playbookWhat it means to run a research lab like an “overgrown garden”How the first two years broke into chapters: research previews, recruiting leading interpretability researchers, then sharpening research directions and building SilicoWhat the OpenAI and Hugging Face incident revealed about reward hacking, chained exploits, and agent swarmsWhy Silico sits one level of abstraction above Claude Code and CodexWhy Goodfire sells top-down to enterprises, with contracts one to two orders of magnitude larger than Eric’s SaaS dealsWho buys interpretability: frontier labs, neolabs, data labeling companies, and enterprises training their own modelsHow Goodfire navigates Anthropic as an investor, partner, and potential competitor

Omni builds business intelligence software that puts exploratory analysis and governed dashboards in a single platform. In a recent episode of UNICORN BUILDERS, we sat down with Colin Zima, Co-Founder of Omni, to learn how the company navigated 18 months to real product-market fit, lost all five of their first trials due to bugs they had learned to work around, and built a demand generation channel where founders still run 10 demos a week at 275 employees.Topics Discussed:How Colin went from being Looker's fourth customer to employee #40, and what watching that company scale to 1,000 people on the way to the Google acquisition taught him about company building stagesWhy the original startup he started with his co-founder never had product-market fit, and what was different about OmniHow Omni found message-market fit in roughly three months but took 18 months to find real product-market fitWhy they lost all five trials at the one-year mark due to bugs they had learned to work around as daily usersHow they won all five deals in the second cohort after two months of product cleanupWhy Omni hired sales reps before the founders finished proving they could sellHow the Looker network gave them the ability to skip validating sales-market fitWhy their third sales rep was in Dublin, with roughly half their early inbound coming from EuropeHow founders doing LinkedIn outreach, with BDRs queuing messages, became a scaled demand channelWhy Colin still personally answers support tickets and runs 10 demos a week with 275 employeesHow Omni became the pink company from a last-minute t-shirt decision at a Snowflake conferenceWhy Colin positions Omni as business intelligence rather than creating a new category like semantic layerHow Omni operates with only 40 to 50 engineers out of 275 employeesWhy Omni has not hired an in-house lawyer for 275 employees

General Magic pursued a vision Marc Porat developed in 1988: a personal handheld computer combining computing, wireless communication, and content. Backed by founding partners including Apple, Sony, and Motorola, the company went public through what Porat describes as a “concept IPO.” In the interview, he puts its value at roughly $800 million. In a recent episode of Unicorn Builders, we sat down with Marc Porat, founder of General Magic, to discuss how the company attracted prominent engineers and major technology partners, why its licensing strategy became a burden on execution, and how going public increased the pressure to deliver. The conversation also covers General Magic’s early work on mobile agents and Porat’s perspective on AI’s promise and risks.Topics Discussed:How Porat developed his vision for a personal, connected handheld computer in 1988How Apple helped bring together the early team, including Bill Atkinson and Andy HertzfeldWhy Apple, Sony, and Motorola became General Magic’s founding partnersHow General Magic paired its operating system and reference hardware platform with partners responsible for manufacturing devicesHow major partners, media coverage, and secrecy generated interest ahead of its “concept IPO”Why public-market financing increased expectations while the company struggled to shipHow a consortium of 16 companies created competing demands that consumed Porat’s timeWhy engineers eventually asked him to stop signing new partnersHow building “heaven for engineers” encouraged creativity but weakened the focus on shippingGeneral Magic’s early work on mobile agents and Porat’s advice for keeping human needs at the center of technology

Building the real-time infrastructure layer for voice AI is a narrow technical bet that turned into a platform play. LiveKit started as an open source project during the pandemic and grew into the stack that powers OpenAI's voice mode for ChatGPT, handling the low-latency network infrastructure that connects users to AI models anywhere in the world.We sat down with Russ d'Sa, co-founder at LiveKit, to learn how an open source project with no commercial intent became the foundational infrastructure for voice AI applications, and what it took to recognize product market fit even when every signal was pointing directly at it.Topics discussed:- Why LiveKit started as an open source project during the pandemic and hit the GitHub top ten repos across all languages within two weeks of launch- The acquisition offer LiveKit received on day one of its first open source release, and why Russ asked for $100 million- How OpenAI signed up for LiveKit in secret using a personal email address, three weeks before reaching out to start a commercial relationship- Why Russ denied product market fit twice despite having a top GitHub repo, enterprise interest from companies like Spotify and Reddit, and an inbound from OpenAI- The difference between stateful and stateless computing, and why the entire software development lifecycle has to be rebuilt for voice AI applications- Why going all in on voice AI took months longer than it should have, and what Russ would do differently- How LiveKit built its full platform stack from network infrastructure to agents framework, evals, inference routing, and observability- Why all of LiveKit's growth to date has been inbound, and what Russ is now building out on the go-to-market side for the first time- How the OpenAI partnership created two separate demand channels: enterprise inbound from the logo halo effect, and developer adoption by association

Tive makes IoT trackers that tell companies exactly where their shipments are and what condition they are in as they move through the global supply chain. The company tracks temperature, humidity, shock, light, and location across 2G through 5G networks, inside trailers, containers, and pallets.In a recent episode of Unicorn Builders, we sat down with Krenar Komoni, CEO and Co-Founder of Tive, to learn how the company reached $100 million in ARR over 11 years by solving a visibility problem most logistics companies were still handling through phone calls, emails, and check calls with truck drivers.Topics Discussed:How Krenar spotted the convergence of LTE IoT, low-cost sensors, and GPS technology before starting TiveWhy the first tracker cost $150 to make and why that could not scaleHow switching to a single-use, sub-$50 tracker changed customer conversationsWhy Krenar got back on the phones with SDRs in 2017 and 2018 and what he learnedHow Tive educated the market for its first seven years and why that work is mostly doneThe near-death moment in 2019 when Tive had $20,000 in the bank and six employeesWhy Krenar stepped back from sales between $5 million and $7 million ARRThe mistake of hiring late-stage sales leaders into an early-stage companyHow Tive evolved from generalist selling to vertical pods and geographic specializationWhat Krenar means by upgrading his CEO brain from V1 to V2 to V3GTM & Technology Adoption Lessons:Time your entry around technology convergence, not just market need: Krenar spent years designing chips for smartphones and watched LTE IoT evolve before starting Tive. He could see that sensors were becoming cheap, that LTE connectivity was getting power-efficient, and that GPS was going to work in small form factors. He started the company knowing it would take years before the economics worked. The product cost structure was the first adoption blocker: The original Tive tracker cost $150 to make and sold for $250. Customers had to return it. That model could not generate the volume or simplicity the market needed. Switching to a single-use tracker at around $50 changed what customers were being asked to do: turn it on, put it in the shipment, and never think about it again.Founder-led SDR work surfaces what customers actually want: In 2017 and 2018, Krenar sat down with a small SDR team and ran cold calls and cold emails himself. That process taught him what the market needed: something cheap, easy, and single-use. That customer intelligence drove the product decision to build a sub-$50 disposable tracker and manufacture at scale in China. ROI stories became the adoption engine: A tracker detecting a temperature spike above 40 degrees Fahrenheit in a truck of strawberries, a diverted shipment worth half a million dollars recovered through location data, pharmaceuticals located on an airport tarmac when the carrier did not know where the pallet was. Hiring a late-stage sales leader into an early-stage company is a specific and recoverable mistake: Krenar stepped in it twice. The issue is not that those leaders are bad. The issue is that a company still figuring out its product and operations does not yet have the foundation for someone who has scaled a business to hundreds of millions. Vertical and geographic specialization unlocked scale: In the early years, every salesperson sold everything to everyone. A person covering EMEA might close a deal in Mexico. At scale that breaks down. Tive now runs vertical pods for food, pharma, and general logistics, with geographic ownership on top. // Sponsors: Front Lines — Silicon Valley's leading Podcast Production Studio. We help B2B tech companies launch, manage, and grow podcasts that drive demand, awareness, and thought leadership. Mention you are a listener and get a 10% discount. www.FrontLines.io/Podcast-as-a-Service