The Digital Diaries Hosted by Peter Woods

orbes 30 Under 30 honouree Tyler Hochman, founder of SafeStop and Four, shares how AI is transforming engineering output, why execution beats ideas, and the three skills every modern founder needs.

Episode Overview
Tyler Hochman started his entrepreneurial journey cutting and selling gems in high school. By his junior year at Stanford, he had launched his first company. Since then he has co-founded SafeStop, a technology platform designed to make police traffic stops safer for both officers and drivers, and Four, an AI solutions architecture firm working with Fortune 500 businesses, sports teams and fashion houses on the data foundations that make AI actually work. Recognised by Forbes as one of the 30 Under 30, Tyler's story is less about the accolades and more about the mindset that earns them: relentless curiosity, thick skin and an obsessive commitment to solving real problems.
In this episode of The Digital Diaries, Tyler shares how AI has changed what is possible for lean founding teams, why virality became SafeStop's biggest challenge rather than its goal, and what he would tell any young founder starting out today.
Ideas are a starting point, not the workTyler treats ideas like a funnel. You need 20 to 50 options before committing to one. The real work is execution, and execution means doing the boring things properly: setting up your CRM, designing scalable architecture and building the foundation before the exciting tools go on top.
How AI has transformed what a small team can achieveA middle-of-the-pack engineer who previously produced 5,000 lines of code per month can now produce 30,000 to 40,000. Tyler argues AI has raised the floor so dramatically that the gap between top and mid-tier talent has narrowed, and lean teams of ten people can now build billion-pound businesses. Every function, including engineering, sales and lead generation, needs to be touched by AI.
SafeStop: when virality becomes the problemSafeStop was built to improve the safety and experience of traffic stops for both drivers and officers. The challenge turned out not to be getting people to want it, but that thousands of people downloaded it in areas where the police departments had not yet partnered with the platform. It is a lesson in being under-prepared for scale that directly informed how Tyler built Four.
Four: the unsexy work that makes AI usefulMost businesses have not set up the data foundations that make AI effective. Four works in the back end, helping organisations ingest, structure and store data correctly so that the AI tools built on top actually deliver insight rather than noise. Tyler's clients include Fortune 500 companies, sports teams and fashion houses. The work is invisible but essential.
Purpose and profit go togetherTyler is direct: purpose drives profit, not the other way around. The clearest example he gives is CPG brands that brought in wellness celebrities to promote alcohol products. The mismatch between the person's values and the brand's purpose was visible to consumers immediately. Authenticity is not a brand strategy, it is a business strategy.
Three skills every modern founder needsThick skin, to take criticism without treating it as a personal attack. Purpose, which does not have to be world-changing but must be genuinely yours. And obsessiveness, which Tyler believes follows naturally once you have found the first two.

Connect with Tyler Hochman
Four: https://www.foreenterprise.comSafeStop: https://www.safetrafficstop.comLinkedIn: https://www.linkedin.com/in/tyler-hochman-83b547130/

Follow The Digital Diaries and share this with a founder or aspiring entrepreneur in your network. Leave a review to help more people find the show.

Enjoyed This Episode?

The Knowledge Economy Has Collapsed: Maeve Ferguson on the IP to Proprietary Data TransitionEpisode Overview
For years, building a course, packaging your expertise and selling your knowledge online was the playbook. Maeve Ferguson says that playbook is finished. Featured in Forbes and founder of Maeve Ferguson Consulting, Maeve is a former financial advisor and private equity operator who spent years building diagnostic and data infrastructure for experts and high-ticket service providers. She has worked with Ryan Levesque's private clients, delivered results for multi-seven-figure businesses globally, and is now helping established experts make what she calls the great IP to PD transition, moving from intellectual property to proprietary data as the last defensible asset in an AI-accelerated world.
In this episode of The Digital Diaries, she explains why knowledge is no longer valuable, what the next 90 days should look like for anyone whose business was built on IP alone, and how a single well-designed diagnostic could be worth hundreds of thousands of pounds.

Why the knowledge economy has collapsedKnowledge that once commanded premium prices is now freely available through AI tools. Maeve does not see this as a threat but as an accelerant. The mediocre will be eliminated. The truly exceptional will thrive. But those sleepwalking through the middle are already being swallowed up without realising it.
The great IP to PD transition explainedIP is what is between your ears. Proprietary data is what gets built because of that IP. Maeve argues the shift from one to the other is not optional: it is already underway. The question is whether experts build the infrastructure to capture and monetise their data now, or start from zero when everyone else has caught up.
Why diagnostic assessments are the infrastructure of this transitionMaeve has been building quiz and diagnostic funnels for seven years. She explains why a well-designed assessment does not just qualify leads. It captures hundreds of behavioural data points per respondent that compound in value over time. A diagnostic her team built for one client generated 60,000 pounds in its first month at a 14.99 price point. Another client's aggregated dataset had a valuation of 14,250,000 pounds.
How data compounds and who is buying itHealth data is roughly six times more valuable than standard data. Forty thousand rows of properly structured health data sold for 340 million dollars. Maeve explains that data aggregated once can be sold to institutional investors, AI companies, and sector-specific buyers repeatedly, across different avenues and use cases.
What the next 90 days look like for an IP-first businessPop the hood. Understand what data you are currently gathering and about whom. Identify the buyers of data in your vertical. Design your diagnostic to output the data points those buyers actually want. Even if data monetisation is not an immediate plan, build with the end in mind today so you are not starting from zero in 12 months.
Using AI as a business building tool, not a threatMaeve uses Whisper Flow with Claude all day across 17 simultaneous work streams for different clients. Her agency now generates personalised proposal websites in minutes after a sales call. Her advice to anyone feeling overwhelmed: start with the biggest bottleneck in your business and just go and play.
Connect with Maeve Ferguson
Website: maeveferguson.comLinkedIn: connect here
Featured in Forbes

Justin Shriber of Terrat explains how AI revenue agents are transforming B2B sales forecasting, deal execution and personalisation, and what the future of the CRO role looks like.

Episode Overview
After nearly three decades leading go-to-market at Oracle, LinkedIn, Siebel and People.ai, Justin Shriber has seen every wave of enterprise software transformation. But he says AI agents feel categorically different — not because of the hype, but because for the first time, a sales rep has a genuine thought partner sitting alongside them in a deal, one that understands context, surfaces risk, identifies best practices from across the entire organisation, and helps the rep execute at a higher level than they could alone.
Justin is CEO and co-founder of Terrat, which is building what he calls the closed loop revenue operating system — an AI-native platform that connects sales execution, forecasting and strategic decision-making into a single compounding system. In this episode of The Digital Diaries, he breaks down exactly how it works, what B2B companies keep getting wrong with AI, and why talent plus hard work will always beat the tool alone.

What makes AI agents genuinely different in salesJustin distinguishes between AI that retrieves data and AI that truly engages as a strategic thought partner. The difference is context — and the engine behind that context is what Terrat calls the revenue graph: a system that aggregates CRM, calls, email, billing and usage data, makes intelligent connections across all of it, and enables natural language questions like why am I losing? and what would my next best move be?
The closed loop revenue operating systemMost sales tools exist in silos. Terrat's thesis is that the real unlock comes from interlocking sales execution with the forecast, and the forecast with strategic decision-making — a closed loop where every cycle makes the system smarter. Justin walks through the three stages: getting pristine signal from the ground, feeding that into an accurate forecast, and using that forecast as the foundation for strategic decisions.
Why CRM projects historically failThe weak link has always been human input — both for populating the system and for designing it. When a CRO sets up sales stages based on gut instinct, the process is built on intuition rather than evidence. Justin shares a vivid case study: Terrat analysed why a customer's EMEA team was losing 27% more deals than other regions, identified that the proof of concept stage was the culprit, and built a data-driven enablement package — with real language from top-performing reps — that gave every rep a proven playbook.
What most B2B companies get wrong with AI personalisation at scaleThree common mistakes: not building the underlying data graph first (producing generic outputs that don't convert), automating fundamentally flawed processes like SDR outreach rather than reinventing the model entirely, and failing to quantify ROI. Justin's alternative to automated SDR outreach: an AI agent that monitors every account continuously, identifies specific buying signals, creates a highly targeted message and deploys it at exactly the right moment — a rifle rather than a shotgun.
The first thing a CRO or CEO should do with AI — and not delegateEvery revenue leader needs a personal OKR: how do we use AI to accelerate growth on a lower cost basis? That productivity equation — current investment vs. output — is the baseline everything else gets measured against. You can't delegate this to a committee.
Where Terrat is heading in five yearsThe platform is expanding beyond sales into customer success, renewal, expansion and ultimately into the CFO's office — enabling what-if financial modelling built directly on live revenue signal rather than assumptions. The long game is becoming the operating system for the entire revenue function.

Resources & People Mentioned
Terret
Justin Shriber on LinkedIn
Mike Gamson

Logan Yonavjak co-raised $85M in institutional capital, built 250+ investor networks, and kept seeing the same problem: great strategy, wrong people. Now she's using behavioural science and AI to measure what actually predicts leadership success — before the stakes get too high.

Episode Overview
Most business failures aren't strategy problems. They're people problems. That's the pattern Logan Yonavjak kept seeing across private equity firms, impact startups, and sustainable finance — and it's the insight that led her to co-found the Founder Readiness Institute.
Logan has co-raised $85M in institutional capital, built a 250+ investor network, advised founders from seed through Series B, and served as a founding member of ANGELS.vc, a women-led angel investing network. With an MBA from Yale School of Management and a Master's in Forestry and Finance from Yale School of the Environment, she brings a rare combination of financial rigour and human systems thinking to one of the most overlooked problems in business.
In this conversation with Peter Woods, Logan unpacks how the Founder Readiness Institute uses behavioural science and people analytics to measure leadership capacity — and why that matters more than ever in an AI-accelerated world.
Key Learnings
Culture eats strategy for breakfast — and people eat culture. Logan's eureka moment came inside a private equity firm where capital was flowing but C-suite misalignment was quietly killing execution. The CEO couldn't take feedback and couldn't make decisive pivots. No assessment tool flagged it. That gap became her mission.
Leadership capacity is how you think, behave and act under pressure over time. The Founder Readiness Institute measures six dimensions including emotional resilience, purposeful agility, coachability and identity flexibility. These aren't soft skills — they're predictive data points for how someone will perform when complexity peaks.
Purposeful agility isn't just speed — it's speed with directionality. Logan distinguishes between a founder who zigzags on instinct and one who pivots with the goal still in sight. It's the difference between reactive and strategic decision making under fire.
Most people live in sympathetic nervous system mode — and it's costing them. High-pressure leadership keeps founders in fight-or-flight. The best leaders learn to shift into parasympathetic states where the neocortex, not the limbic brain, drives decisions. This isn't abstract wellness — it's neuroscience applied to performance.
AI is exposing the right people and the wrong fits simultaneously. Logan believes AI is making leadership assessment more precise, more accessible and less expensive than ever before. Used well, it removes confirmation bias and the halo effect from promotion and hiring decisions — two of the biggest causes of the 40-50% leadership failure rate.
Only 2-3% of VC funding reaches women and minorities. As a founding member of ANGELS.vc, Logan is working to shift that — not through quotas, but by introducing objective people analytics into investment decision making so that gut feel and warm introductions stop being the dominant filter.

Brendan Norman of Classify explains why contextual intelligence beats raw data, how the agentic web is reshaping advertising, and what digital trust really means in 2025.
Why Context Beats Data: Brendan Norman on the Agentic Web, AI Advertising and Digital TrustEpisode Overview
Most of the conversation around AI focuses on data — who someone is, what they've clicked, what they've bought. But Brendan Norman, Co-Founder and CEO of Classify, argues we're missing something more fundamental: context. Knowing who someone is means nothing if you don't understand where they are mentally when you reach them.
In this episode of The Digital Diaries, Brendan draws on a decade at the intersection of ad tech, platform strategy and go-to-market — from founding client partner at Facebook's Audience Network to strategic supply leadership at Unity Technologies — to explain how Classify is building the contextual intelligence layer for the next era of the internet: the agentic web.

What You'll Learn in This Episode
Why context is more valuable than dataData tells you who someone is. Context tells you whether this is the right moment to reach them. Brendan explains why combining behavioural signals with real-time semantic understanding of content is the difference between an ad that builds trust and one that destroys it — with vivid examples from a Super Bowl ad campaign that went badly wrong.
What Facebook taught him about digital trustFrom his years building the Facebook Audience Network pre-Cambridge Analytica, Brendan shares the three pillars that every ad platform needs to get right: advertiser ROI, publisher monetisation, and user experience. Most platforms treat user experience as an afterthought. That's where trust erodes.
What the agentic web actually isBrendan gives one of the clearest explanations of the agentic web you'll hear: AI agents crawling the web to retrieve and synthesise information in real time, agents communicating with other agents via backend protocols (A2A), and what all of this means for how content is consumed, ranked and monetised. A process that once took two to four weeks of human effort is now running in under 30 seconds.
How advertising integrates into AI-powered workflowsAgentic attention is the next frontier of monetisation — and it's barely been touched yet. Brendan explains how Classify is working on surfacing contextual signals within the backend files that AI agents consume, so that advertising can be inserted in ways that are relevant and non-disruptive, rather than keyword-triggered and tone-deaf.
Fraud, brand safety and the trust layerAs AI agents scale, so does the risk of fraudulent bot traffic inflating impressions, misplaced ads damaging brands, and broken user experiences. Brendan explains how Classify detects invalid traffic, validates impressions and, critically, builds its entire contextual layer without cookies or any PII — focusing purely on content-level intelligence.

Practical AI advice for overwhelmed founders and operatorsDon't hand over your API keys. Do carve out time to play. Brendan shares how rebuilding the Classify website himself using Bolt (built on Claude/Anthropic models) became the catalyst for a much deeper understanding of React, Tailwind, GitHub and modern development. The lesson: use AI as a coach and a learning accelerator, not a replacement.

Books & Resources Mentioned
Bolt — front-end AI website builder (used by Brendan to rebuild Classify's site)
Claude / Anthropic models — embedded in Bolt; also discussed as a writing and coding tool
Cursor — AI coding tool mentioned for developers
Claude Code — mentioned as tooling for building with AI

Connect with Brendan Norman
Company: Classify
LinkedIn: https://www.linkedin.com/in/brendannorman/