High Stakes FM

Sara Maldon is the founder of Inflect, a company that helps organizations measure the value they create with AI internally. Before Inflect, she spent two and a half years as the AI transformation leader at a scale-up of around 350 employees. She joined when AI was prohibited and built the function from scratch: tooling, governance, roadmap, employee enablement, and the measurement layer that became the seed of Inflect.

In this episode, we discuss:

1. Why Sara chose not to treat "time saved" as an AI KPI, and why saved hours often never show up in revenue

2. How redesigning the HR onboarding journey drove AI adoption from day zero

3. What goes into Inflect's AI index: governance, quantity and quality of output, and environment

4. The survey question that shows whether an AI transformation is spreading or stuck with a few individuals

5. How AI ROI can compare across departments, from engineering token spend to HR time to hire

6. How Sara runs a B2B company as a solo founder without building it alone

7. How she decided to leave a great job, and why the decision is rarely a single moment

8. Her advice for aspiring founders who do not have an idea yet

Chapters:

00:00 Introducing Sara Maldon and Inflect

00:41 What Inflect does

01:17 How building an AI transformation from scratch sparked the idea

03:30 What moves the needle, and why time saved is the wrong KPI

06:52 Inside Inflect's AI index

10:42 The survey question that reveals the motivation gap

13:16 Why non-AI questions matter for AI success

14:27 The competitive landscape

16:26 Running a B2B company as a solo founder

19:34 Deciding to leave a full-time job

25:38 Building before making the leap

27:24 Where measuring AI value is heading

30:52 Advice for aspiring founders with no idea yet

33:36 Wrap-up

Where to find Sara Maldon:

LinkedIn: https://www.linkedin.com/in/sara-maldon/

Inflect: https://www.linkedin.com/company/inflect-hq/

Where to find Gabriela Naumnik:

LinkedIn: https://www.linkedin.com/in/gabrielanaumnik/

X: https://x.com/NaumnikGabriela

Website: https://www.gabriela-naumnik.com/

Also on Apple Podcasts: https://podcasts.apple.com/us/podcast/high-stakes-fm/id1896850167

And YouTube Podcasts:

https://www.youtube.com/watch?v=Wr-ZP5BFAfc&t=4s

For sponsorship inquiries, DM Gabriela Naumnik on LinkedIn: https://www.linkedin.com/in/gabrielanaumnik/

William Lindholm is the founder of Daymaker, the company behind "cold caking," which means sending a prospect a custom cake with a QR code on it instead of another cold email. He's 20, Norwegian, and moved his company from Oslo to San Francisco. Daymaker started as an HR tool that automated birthday cakes for employees and scaled to $150K ARR in about five months, then hit a wall when he discovered American companies don't celebrate employee birthdays the way Norwegian ones do. The pivot went viral, pulling 300 million views in a couple of weeks, and the company now runs on a network of 100+ bakeries across the US, Europe, India, and Australia.In our in-depth conversation, we discuss:

• How cold caking works, what it converts at, and the two metrics Daymaker actually tracks
• How job seekers are putting their CVs on cakes to reach frontier AI labs, and the candidate who landed a job at Virio this way
• The 1,000-cake logistics disaster across 50 cities that burned him out and became the company's biggest lesson
• How Daymaker uses AI agents to find, negotiate with, and onboard bakeries with almost no human involvement
• Why he thinks physical marketing, not cake, is the billion-dollar business
• The product they're launching: an agent that reads a prospect's digital footprint and picks the gift
• Why he'd tell any new founder to sell first and build later

Where to find William Lindholm:• LinkedIn:   / william-daymaker  • X: https://x.com/daymakerguy• Daymaker: https://daymaker.comWhere to find Gabriela Naumnik:• LinkedIn:   / gabrielanaumnik  • X: https://x.com/NaumnikGabriela• Website: https://www.gabriela-naumnik.com/Where to find High Stakes:• Spotify: https://open.spotify.com/show/033pQlV...• Apple Podcasts: https://podcasts.apple.com/us/podcast...In this episode, we cover:(00:00) Introduction(00:24) What cold caking actually is(00:57) From HR software to cakes: the origin story(01:45) Correcting the record on $120K/month(02:41) Go-to-market: why the product markets itself(03:15) VC cakes vs. the customers who actually pay(04:24) The two kinds of customers: tech and white collar(05:09) People are putting their CVs on cakes(06:04) Companies sending cakes to candidates they want to hire(06:29) The only two metrics Daymaker tracks(07:24) Copycats and inviting competition(07:56) 100+ bakeries and the agent running deliveries(09:37) The 1,000-cake disaster in 50 cities(11:07) Why physical marketing becomes a billion-dollar category(12:45) The new product: agentic gifting, worldwide(13:37) Running a company with three people(14:31) Who Daymaker is hiring, and what they look for(15:20) AI agents that source and negotiate with bakeries(17:23) What the hyper-personalized gifting space looks like(18:46) Advice for anyone starting something nowReferenced:• Daymaker: https://daymaker.com• Coldcaking explainer: https://daymaker.com/coldcaking• Virio: https://www.virio.ai/• Forget pitch decks, you need pitch cakes (Sifted): https://sifted.eu/articles/forget-pit...• Let Them Pitch Cake (Inc.): https://www.inc.com/mariapaula-gonzal...• Cake it till you make it (The National): https://www.thenationalnews.com/news/...• Rippling: https://rippling.com• Ramp: https://ramp.com• Brex: https://brex.com• University of Edinburgh: https://www.ed.ac.uk• OSINT (open-source intelligence): https://en.wikipedia.org/wiki/Open-so...For sponsorship inquiries, DM Gabriela Naumnik on LinkedIn:   / gabrielanaumnik  

Humans tell you what you want to hear. So do the LLMs trained to be nice to you. For Tom Charman, that is the trap most synthetic research walks into, and the problem he built Blok to solve.

Tom is the co-founder of Blok, a simulator for product teams. As he describes it, building is getting faster and cheaper, so the hard question is no longer how to build. It is whether you should build the thing at all. Blok puts agents that represent real customer segments directly inside your product flow, then shows step by step how those users behave, where they feel friction, and where they drop off.

Tom and Gabriela Naumnik get into the thesis underneath the product: that experimentation is shifting from reactive to proactive, and that most companies will never have the data to do it the old way.

What Tom covers in this episode:

- Bottom-up vs top-down, and why agents interacting with the product beat a population answering questions
- The 87% fidelity benchmark Blok reports and what it takes to trust simulated results
- Why regulated industries in finance, healthcare, and government came first for Blok
- How discovery can drop from four to six months to one or two days
- Why adaptive interfaces could make simulation the only path to statistical significance for 99.9% of companies

For founders, product leaders, and operators deciding what to build next.

highstakes.fm

What if achieving SOC2, HIPAA, or ISO compliance took 5 hours instead of 5 months?

Claudio Fuentes, COO and co-founder of Comp AI, breaks down how they're automating the entire compliance process - and why going open source was their biggest strategic bet.

In this episode, we go straight to the contrarian calls:
- Why Comp AI ditched the black-box approach and went fully open source
- How they compress hundreds of hours of compliance paperwork into 5-6 hours using AI
- The Henrik persona: satirical marketing that cut through a notoriously dull industry
- What happens to compliance jobs as automation takes over routine tasks
- How startups can use fast compliance to unlock enterprise deals

Built for startup founders, VCs, operators, and anyone in tech who wants to understand how the best builders actually think.

highstakes.fm