Every investor who passed on Sage passed for the same reason: "You're awesome, your team is awesome. Why are you building hardware? You could get a great clean seed check if you just build software." Six years and $125 million later, the hardware is the reason Sage has no churn and 180% net revenue retention.

Raj Mehra is the co-founder and CEO of Sage, which replaced the nurse call system, a technology essentially unchanged for 40 years, with IoT devices, mobile software for caregivers, and a physical AI layer that went from proof of concept to millions in ARR in six months. Before Sage he failed at a healthcare startup at 24, built the first health system deployment at Palantir, and was an early hire at Cedar.

Marco DeMeireles sits down with him at Ai Fiori in New York to walk through every round Sage raised and name exactly what the room got wrong each time. It's the rare founder conversation where the fundraising story is the argument, not the victory lap.

In this episode:

• Why every investor in the seed round told him to drop the hardware, and how that hardware became the moat

• Raj’s formidable failure at Palantir

• Why Raj’s last company, Cedar, stalled

• Sage's role in the evolution of physical AI

• How Raj tested his co-founder’s for fit

• The way Sage’s machine learning models were trained

• The differentiated work Maveron did at the Series A

• How to use debt to creatively fund growth without blowing up the company

Undercovered is the show where founders building mission-critical companies get the conversation before the world catches up. New episodes weekly on YouTube, Spotify, and Apple. Subscribe so you hear these stories first.

Timestamps:

00:00 Intro

01:59 How he ended up building for senior care

07:10 What Palantir and Cedar each taught him

09:15 His most formidable failure

16:12 The right thesis at the wrong time

19:21 Physical AI and where Sage fits

21:32 The first product, and why nurse call isn't sexy

23:07 Selling with a V0 product and a limited hardware run

25:05 Three co-founders and how they tested each other

28:14 The values they wrote in McCarren Park

33:36 What it costs a 100-bed facility to adopt

34:47 Where the AI actually lives in the product

44:47 Why the hardware made investors nervous

46:01 The investor who toured facilities with her mom

47:43 Using debt instead of more equity

50:49 What investors still miss about this sector

For 12 years, Alex Konrad decided which founders and investors the startup world paid attention to. He ran the Midas List and the Cloud 100 at Forbes, wrote cover stories on Stripe, Zoom, and Snowflake, and met Canva's Melanie Perkins when she was a line item on a rising stars list. Then he left to build Upstarts Media, and found out what it actually costs to be on the other side of the table.

Over lunch at Ai Fiori, Marco DeMeireles sits down with a journalist who is now a bootstrapped founder trying to grow a media business in a moment when every piece of advice on his timeline says raise big or die. Alex is candid about the parts nobody writes about: charging readers the price of a sandwich, wanting more complaints in his comments, and the discomfort of hiring people after leaving a place that profited from his own work.

The conversation also runs the other direction. Alex turns the questions back on Marco about pricing, diligence, and how investors know if something is medicine or a vitamin.

In this episode:

- Why Forbes capped Midas List credit at two people, and how firms tried to game it anyway

- What it takes to make a reporter open your email, and why success alone doesn't do it

- The red flag that makes Alex walk away from a founder before he writes a word

- How he found a bootstrapped company with hundreds of millions in revenue that nobody was covering

- Why he's paranoid about AI-manufactured narratives showing up as fact

- The beat he thinks deserves a full-time reporter before it becomes obvious

Subscribe for new episodes every week.

Undercovered is on YouTube, Spotify, and Apple Podcasts.

Timestamps:

00:00 Intro

01:47 The two founders who come to mind first

03:09 Why he left Forbes to build Upstarts

05:18 Bootstrapping while everyone says raise big

07:12 Apathy is the killer

07:24 The paywall, and charging the price of a sandwich

08:24 How he decides who to cover

09:32 Why he wants more complaints

10:34 Medicine or a vitamin?

12:23 Owning your work after leaving a brand

14:47 How the AI boom compares to the ones he covered before

19:11 Meeting Melanie Perkins on a rising stars list

19:47 How the Midas List data actually worked

21:02 VC egos and who gets credit for a deal

23:46 Why Forbes capped Midas credit at two people

24:13 Rural Pennsylvania and a board-game savant

28:30 The signals that make him lean in

32:43 Non-obvious red flags

34:55 Why he’s paranoid about AI-manufactured narratives

41:32 Quickfire round

Undercovered is a new show from Ansa Capital Co-Founder Marco DeMeireles, built around one idea: the founders worth talking to are often the ones nobody's heard of yet. Each episode is filmed over a real meal at one of New York's finest restaurants, with a founder building something in defense, cybersecurity, healthcare IT, or another market most people ignore until it breaks.

In this episode, Adam Delehanty writes the words other people take credit for. Half his family are writers, lawyers, and journalists. The other half includes a mother who is a social worker and therapist. He describes the job he built out of that inheritance as journalism times therapy: journalism to get the story right, therapy to hold space long enough for a founder to actually deliver it.

Ghost is bootstrapped, six years old, and deliberately small. A handful of people, six to twelve clients at a time, a few books a year. He has turned down the obvious growth path more than once, including the one where he takes venture money and builds an army, because he would rather do the work than manage people doing it.

The books are the part most people have not heard about. Hinge, Duolingo, DocuSign, Crunchyroll, and one he is not allowed to name. His premise is that companies are religions and religions need a Bible, something on paper a company can orient around instead of a galaxy of Slack threads and Notion pages. To write one he will interview fifty or seventy-five people to find the gap between what a founder believes the culture is and what it actually is.

Marco and Adam get into the deltas that keep showing up in those interviews, why he refuses to fill the feed, how he actually uses Claude, the upside-down internet of people who can never post, why he thinks tech reads further right online than it is in private, and the one sentence he wrote that he is contractually unable to claim.

Chapters

• 00:00 Make something so compelling the world can't ignore it

• 02:00 Why founders shouldn't just yap about the product

• 06:59 Ghostwriting as journalism times therapy

• 07:53 Bootstrapping, and why agencies shouldn't take venture money

• 09:14 Companies are religions and religions need a Bible

• 10:04 The delta between the culture a founder describes and the real one

• 12:46 70% truth, 30% aspirational

• 13:15 How the first culture book happened

• 14:37 Why the original founder is the best culture tool you have

• 18:00 The internet we see and the upside-down internet

• 20:04 What happens to tech journalism when every interview is friendly

• 22:12 Pricing writing when the floor is falling out

• 25:24 AI like a sandwich: human, AI, human

• 28:43 What it says about you when you publish slop

• 34:45 What VCs get wrong about content

• 44:47 Finding the mission in a founder's own story

• 50:10 What young writers should do instead of taking the shortcut

• 51:22 The hottest take: we can all tell

• 52:52 The writer nobody knows, and living an original life

• 55:24 Why he thinks tech is more centrist than the feed suggests

• 55:58 The famous sentence he can't take credit for

Undercovered is a show from Ansa co-founder Marco DeMeireles, built around one idea: the founders worth talking to are the ones nobody's heard of yet. Each episode is filmed over a real meal at one of New York's finest restaurants, with a founder building something in a market most people ignore until it breaks.

In this episode, Zach Cohen was an investing partner at Andreessen Horowitz, working largely in consumer and consumer health, before resigning to build Ease Health. He flew out to tell his boss he was leaving, and the response was to ask for the deck. a16z led a term sheet inside 24 hours, and the full $41M Series A came together over roughly 72. He has spent the ten months since building three overlapping systems of record at once, CRM, EHR, and revenue cycle, for behavioral health providers who currently run their business across a dozen disconnected vendors.

Marco and Zach get into why Zach thinks inpatient psychiatric care, much of it still on paper, is one of the most undercovered problems in healthcare, why he refuses to lead sales pitches with headcount reduction, why he built in New York instead of San Francisco, and what it's like managing four concurrent product roadmaps ten months into a company growing 7 to 8x.

Chapters

• 00:00 The 24-hour term sheet

• 02:10 What it means when the firm you are leaving leads your round

• 04:14 From consumer investing to behavioral health

• 07:10 Moonlighting in ambulatory clinics for six months before starting

• 09:18 Why behavioral health isn't actually crowded

• 12:25 What AI-native software can do that legacy EMRs can't

• 15:03 Why he may one day give ambient scribing away for free

• 16:26 Why build three systems of record at once

• 21:51 The "heart surgery" sales pitch

• 24:37 Verticals becoming diagonals

• 27:56 Why he doesn't lead with labor replacement

• 33:51 Choosing New York over San Francisco

• 38:16 Hiring the first five, before the round even closed

• 43:04 What's in the "fear doc"

• 54:15 Growing 7 to 8x, and accelerating into next year

• 57:15 The undercovered problem: inpatient psych still on paper

• 59:53 His billboard for a billion people

Undercovered is a new show from Ansa Capital Co-Founder Marco DeMeireles, built around one idea: the founders worth talking to are often the ones nobody's heard of yet. Each episode is filmed over a real meal at one of New York's finest restaurants, with a founder building something in defense, cybersecurity, healthcare IT, or another market most people ignore until it breaks.

In this episode, Taylor Offer is the first man in four generations of his family not to work in the movie business. His great grandfather, grandfather, uncle, and father all ran film. Then the industry went digital, his dad was a studio projectionist in his 50s with no other skillset, and the family found out what it costs when a technology decides you're no longer necessary.

He says the phrase "at the mercy of" three times in this conversation, about three different losses. His father's job. Facebook, which took every dollar of Feat's margin until 5:1 ROAS became 1:1. And the co-founder his board made him fire, a best friend of seven years he spoke to more than his own family.

Atrios is his third answer to the same problem: stop renting the distribution channel and become it. This time the channel is other people's trust.

Marco and Taylor get into the Google Earth search that found the Vine house, the Logan Paul sock deal that pulled in 20,000 customers, what a private equity sale actually looks like from the inside, cutting a third of the team, and what he thinks the most under covered problem in tech is.

Chapters

• 00:00 "You die when you stop believing you could do it"
• 02:21 Four generations in the movie business
• 04:41 "I will never be at the mercy of a job"
• 05:13 The ten-dollar coffee, and what humans still get paid for
• 06:29 Building Feat to $50 million
• 07:29 A hundred cold calls a day, and an app called Vine
• 08:04 Finding the influencer house on Google Earth
• 08:46 The Logan Paul sock deal
• 10:17 Scaling on Facebook
• 11:12 When five-to-one ROAS became one-to-one
• 11:27 Deciding to sell
• 12:48 The accidental discovery of the tastemaker
• 13:08 What founders don't know about a PE sale
• 14:10 Cutting a third of the team
• 17:50 Why Atrios exists
• 18:13 Is the SDR role dead
• 20:12 A CRM for tastemakers
• 23:41 Whether venture is the right partner
• 26:51 Building without a co-founder
• 28:48 The board made him fire his best friend
• 33:44 "More wars are lost by loss of hope than loss of blood"
• 35:17 The under covered problem: nobody knows how to sell
• 37:54 The people earning $10-20k a month
• 54:10 Where to find Taylor