Teahose.
SIGN IN
NEW HERE — WHAT TEAHOSE DOES
We read the entire AI & tech firehose — so you don't have to.
PODPodcastsAll-In, No Priors, Acquired…
NEWNewslettersStratechery, Newcomer…
PAPPapersPhysical AI research
PHProduct Huntdaily launches
VCInvestor ScoutSequoia, a16z, Benchmark…
CLAUDE DISTILLS →
7 reads, 30 sec each — free, 6 AM ET.
+ a live graph of the companies, people & themes underneath.
HOME/SOURCERY NEWSLETTER/ICYMI: Sourcery Live in the Sout…
NEWS
// NEWSLETTER ISSUE
SOURCERY NEWSLETTER

ICYMI: Sourcery Live in the South of France

DATE October 3, 2026SOURCE SOURCERY NEWSLETTERPARTICIPANTS MOLLY O'SHEA
// SUMMARY

1. Key Themes

Theme: The AI buildout is shifting from compute to interconnect, power, and data movement (photonics)

Copper is hitting its limits as models outgrow single GPUs

Thema's CTO, a former NVIDIA photonics engineer, argues that model scale is forcing the move from electrical to optical interconnect.

"The models have gotten so large that they no longer fit on a single GPU." "Copper is running out of steam."

Interconnect is roughly half of an AI factory

The non-GPU half of the stack is now a major constraint and a major opportunity.

"If you look at an AI factory today, 50% of that is GPUs, but the other 50% is the technology to interconnect these GPUs."

Constraints extend beyond chips to power and data movement

"We go inside the race to scale photonics manufacturing as copper, power & data movement become increasingly important constraints on AI factories."

Memory is a concentrated chokepoint

"There's only 3 companies in the world that produce memory, which is 95% of the market: Micron, SK Hynix, & Samsung." "Why are memory prices up 700% this year?"

Hyperscalers taking over inventory is cited as a driver.


Theme: Europe's defense rearmament is a manufacturing and scale problem, not just a capital problem

Money is available; industrial capacity is not

"Europe is rearming after decades of underinvestment, but capital alone doesn't solve the industrial problem." "Prototypes don't fight wars."

Delivery timelines mismatch modern warfare

"Some contracts take 4–5 years to deliver, while modern warfare requires systems…" (truncated in the source tweet)

The opportunity framing is large

"Could Europe build a $500B Defense company?"

The "Tech or Industrial: Who wins in Defense?" and "How autonomous warfare is reshaping the supply chain" segments suggest the winners will blend software with manufacturing at volume.


Theme: Physical-economy infrastructure companies are scaling at software-like speed

Logistics as an AI/robotics platform

"Stord is crossing $1B in annual revenue after 10Xing the business in four years. The company now handles roughly 100M packages a year, ~$20B in GMV & 8B operational data points as it makes a major push into robotics and AI."

Global sales at extreme velocity

"Deel went from $1M to $100M in ARR in under 20 months, then from $100M to $1B in three years."

Series framing

"Sourcery Live goes inside some of the biggest shifts & hottest takes happening across AI infrastructure, defense, logistics, photonics, semiconductors & global company building."


Theme: Semiconductor roadmaps are moving beyond today's AI paradigm

Long-horizon R&D view

imec's Steven Latré predicts a major leap beyond current AI:

"I think we're gonna, five to 10 years from now, we're gonna laugh at how old-fashioned AI was today."

Hardware is increasingly the leverage point

Segments such as "The end of Moore's Law" and "Is hardware the new software?" point to scaling limits pushing innovation into memory, photonics, and architecture.

"Almost every chip in the world today has been touched by imec in some way."


2. Contrarian Perspectives

Founders aren't selling enough

Deel's CRO positions under-selling as the common failure mode, countering the narrative that product or funding is the main bottleneck.

"Hot take: People need to sell more" / "Hot take: Why founders need to sell more" "We break down the sales machine behind the company, building a global organization from day one & why Shuo thinks founders simply aren't selling enough."

Evidence: Deel's growth from $1M to $100M in under 20 months and $100M to $1B in three years, attributed to treating sales as a system ("Why sales is really about systems").

AI and robotics will become commodities

Stord's CEO, while pushing hard into robotics and AI, argues that the technology itself won't be the moat. The timestamped segment is titled:

"Hot take: Why AI and robotics will become commodities"

Combined with his Thiel Fellowship lesson ("build a monopoly") and "Why owning the whole supply chain matters," the implication is that defensibility comes from owning the operating network and data, not the underlying tech.

Choice is an illusion / Europe isn't as far behind as assumed

Two segments challenge conventional wisdom: the defense segment "'Choice is an Illusion'" and the Thema segment "The Biggest misconceptions about European AI," plus "Is Europe actually ready for this?" The framing suggests Europe has underappreciated capability, but the bottleneck is manufacturing and volume rather than ideas.

"Prototypes don't fight wars."


3. Companies Identified

Deel

  • Description: Global people/payroll platform
  • Why mentioned: Case study in hypergrowth through systematized sales and global-from-day-one design
  • Quotes: "Deel went from $1M to $100M in ARR in under 20 months, then from $100M to $1B in three years." / "Trusted by more than 35,000 fast-growing companies." Customers range "From ElevenLabs to Chanel." Valuation cited: "$17.3B Deel."

Stord

  • Description: Logistics and fulfillment company pushing into robotics and AI
  • Why mentioned: Positioned as a challenger to Amazon Prime with owned supply chain
  • Quotes: "The $3B Company Trying to Beat Amazon Prime" / "100M packages/year, ~$20B in GMV, $1B+ revenue this year, after 10Xing in 4 years, 8B operational data points."

Thema

  • Description: European photonics manufacturing startup
  • Why mentioned: Building photonics capacity as AI interconnect becomes a bottleneck
  • Quotes: "Thema is building photonics manufacturing capacity in Europe."

imec

  • Description: Leading semiconductor R&D organization
  • Why mentioned: Developing chip technologies years ahead of the market
  • Quotes: "One of the most important semiconductor organizations most people have never heard of." / "Almost every chip in the world today has been touched by imec in some way."

Onodrim

  • Description: European defense company (CEO Aistis Šimaitis)
  • Why mentioned: Participant in the European defense buildout discussion
  • Quotes: Aistis Šimaitis is "CEO & Co founder of Onodrim."

Theon International / EFA Group

  • Description: Defense and night-vision/optronics group
  • Why mentioned: Incumbent industrial perspective on European defense
  • Quotes: George Hadjiminas is "Chief Strategy Officer - Theon International & EFA Group."

Strike Capital

  • Description: Venture investor and host sponsor of the series
  • Why mentioned: Investor perspective across defense and semiconductors
  • Quotes: "Courtesy of our friends at Strike Capital."

NVIDIA

  • Description: AI chip leader
  • Why mentioned: Origin of Thema's CTO's photonics expertise; described as reshaping supply chains
  • Quotes: "Why NVIDIA broke the supply chain" / Yannick "left NVIDIA after helping take its silicon photonics technology from scratch to product-ready maturity."

Palantir

  • Description: Data analytics/defense software company
  • Why mentioned: Source of operating lessons for European defense builders
  • Quotes: "Lessons from Alex Karp's Palantir" / "Palantir's 2nd employee in Europe, Aistis Šimaitis…"

Micron, SK Hynix, Samsung

  • Description: The dominant memory producers
  • Why mentioned: Concentration explaining memory price spikes
  • Quotes: "There's only 3 companies in the world that produce memory, which is 95% of the market."

Dyson

  • Description: Engineering-driven consumer products company
  • Why mentioned: Inspiration for Shuo Wang's founder philosophy
  • Quotes: "My first company, I was inspired by James Dyson." / "He's just obsessed with solving a very specific engineering problem and then making that product perfect."

4. People Identified

Shuo Wang

  • Description: Co-Founder & CRO of Deel
  • Why mentioned: Architect of Deel's sales machine
  • Quotes: "I'm going to be client #1." / "Why sales is really about systems."

Sean Henry

  • Description: Founder & CEO of Stord, Thiel Fellow at 19
  • Why mentioned: Scaled Stord to $1B+ revenue; monopoly-building mindset
  • Quotes: "I think they really impress upon you: what is your right to win and essentially build the biggest company possible, i.e., build a monopoly over time."

Yannick De Koninck

  • Description: CTO of Thema, ex-NVIDIA
  • Why mentioned: Deep expertise in silicon photonics
  • Quotes: "Copper is running out of steam."

Herwig Van Hove

  • Description: Founder & CEO of Thema
  • Why mentioned: Leading the effort to build European photonics manufacturing
  • Quotes: Segment "From Investor to Operator."

Aistis Šimaitis

  • Description: CEO & Co-founder of Onodrim; early Palantir Europe employee
  • Why mentioned: Operating philosophy drawn from Alex Karp
  • Quotes: "Focus on being important & valuable, & the money will come."

George Hadjiminas

  • Description: Chief Strategy Officer, Theon International & EFA Group
  • Why mentioned: Industrial defense perspective
  • Quotes: "Europe is rearming after decades of underinvestment."

Bruno Wejchert

  • Description: Strike Capital investor
  • Why mentioned: Investor lens on defense
  • Quotes: "Europe's Biggest Defense white space."

Adam Chambers

  • Description: Strike Capital investor
  • Why mentioned: Co-interviewer on semiconductors with imec
  • Quotes: "Is there a limit to how many chip companies can exist?"

Steven Latré

  • Description: VP of AI at imec
  • Why mentioned: Long-horizon semiconductor and AI forecasts
  • Quotes: "Five to 10 years from now, we're gonna laugh at how old-fashioned AI was today."

Alex Karp

  • Description: Palantir CEO
  • Why mentioned: Mentor-style influence on defense operators
  • Quotes: "Lessons from Alex Karp's Palantir."

James Dyson

  • Description: Founder of Dyson
  • Why mentioned: Product-obsession inspiration
  • Quotes: "He's just obsessed with solving a very specific engineering problem and then making that product perfect."

Alex (Deel co-founder)

  • Description: Deel co-founder paired with Shuo
  • Why mentioned: Role-division model
  • Quotes: "How Shou and Alex divide the roles."

5. Operating Insights

Treat sales as a system, not heroics

Deel's growth is attributed to repeatable process, including a remote sales team and global go-to-market from the start.

"Why sales is really about systems." / "How to build a winning remote sales team." / "Why Deel went global from day one."

Be your own client #1

Wang's product philosophy: use what you build and obsess over a specific problem.

"I'm going to be client #1." "Obsessed with solving a very specific engineering problem and then making that product perfect."

Optimize for importance and value, not just revenue

Šimaitis's Palantir-derived principle for building durable companies, especially in defense.

"Focus on being important & valuable, & the money will come." "How am I creating value? Why we are becoming more important as a company for the world?"


6. Overlooked Insights

Supply constraints are a real gating factor for photonics commercialization

The Thema segment flags commercial timelines and supply limits, a reminder that even in a hot category, production capacity gates revenue.

"Commercial progress, timelines and supply constraints."

Data center backlash differs between Europe and the US

A potentially important political and siting variable for infrastructure investors.

"Data center backlash in Europe vs. the US."