🍍 Gabriel Jarrosson

Lobster Capital

Managing Partner

Leonis Investissement

Investing in YC startups. $36M deployed since 2020.

Chairman

Founder & CEO

Piña Colada

US stocks recommendation newsletter
Company sold in 2023 (acquired by Agora France)

Founder & CEO [acquired by Agora France]

albert académie

Online e-learning platform. 
Company sold in 2023 (acquired by P2KR)

Co-Founder & CEO [acquired by P2KR]

Le Vin de France

Marketplace for french wine growers.
Company sold in 2016 (acquired by Les Grappes).

Founder & CEO [acquired by Les Grappes]

🔩 Computer Science Engineer
💪🏻 7x founder with 3 exits (4 failures, learned some stuff). Bootstrapped to $1M ARR with 0 employees. 
🦞 VC @ Lobster Capital: Funding the best YC startups to change the world.

Raising money in Germany for his AI startup? Nearly impossible.

Raising several million in a week in Silicon Valley? Easy.

Kevin Wu built a voice AI company in Germany. Couldn't raise money. "No early stage capital for technical founders with unproven businesses."

Gets into YC, moves to San Francisco. Revenue doubles in two months. Raises $4.7M in a week with extra offers.

I’m sure the difference isn't talent… Wu clearly knows what he's doing - 10,000 calls handled daily, 90% customer satisfaction.

The difference is mindset. European VCs want proof before they invest and Silicon Valley VCs invest to get proof.

This means I would effectively discount anyone in Europe who says they do “early stage” to “sort-of early stage”.

I watched this pattern repeat - European founders who moved to the US consistently raised more, grew faster, and scaled bigger than those who stayed.

The problem compounds at later stages. When you need growth capital, European VCs become even more conservative. They want perfect metrics and guaranteed outcomes.

Meanwhile, US growth funds write $50M checks based on momentum and market opportunity.

That perception gap explains everything. Same founder, same product, different location - completely different investor reception.

I'm not anti-European, the talent is really there but I'm pro-results. And the results consistently favor founders who relocate to ecosystems that bet on potential over proof. 

Thanks in advance for all the exceptions in the comments!

Gabriel Jarrosson

VC @ Lobster Capital 🦞🧢 Investing in the top 2% of YC startups | Founder with 3 exits

YC just added another unicorn to their portfolio - Eight Sleep hit $1B valuation.

Eight Sleep spent their first five years trying to fit in. They positioned themselves like every other startup in their space - mattress-in-a-box companies, IoT devices, wearables.

Classic "teenager syndrome" i.e. copying what seems to work rather than defining who they were.

But there was little success in being compared to mattress companies. So they had to create their own category.

They stopped calling themselves a mattress company and pioneered "sleep fitness." Not just a product pivot, but a complete repositioning around the mission of optimizing performance through sleep.

This built a movement and rode the trend of ‘sleep hygiene’ from wellness etc.

Key insight from co-founder Alex Zatarain -

"Your goal as a brand builder becomes to make that category a thing. You're building a movement. You're bringing people into your church of sleep fitness and evangelizing this lifestyle ahead of evangelizing your products."

We see the same across our most successful YC companies at Lobster Capital. The biggest winners are making new categories.

Stop trying to fit into existing boxes. Build your own category and make it matter.

🚨 Another YC unicorn minted 🚨

Eight Sleep (YC S15) just raised $100M at a $1B valuation—but the real story isn’t funding.
It’s how they escaped the “mattress trap” by creating a new category: Sleep Fitness.

I break down:
YC’s category creation playbook
How data + execution built a moat
Where Eight Sleep goes next (AI + healthcare)

🎥 Full episode here ⬇️
https://lnkd.in/eh448arg

YC just told Apple to stop killing startups.

The world's most successful accelerator filed an amicus brief against Apple's App Store policies.

The 30% “Apple Tax” can easily be the difference between a company that can scale and one that is perpetually struggling to stay afloat.

At Lobster Capital, I've watched promising mobile startups die not from lack of users, but from impossible unit economics. When Apple takes 30% and you're already running thin margins, the math simply doesn't work.

YC admits they've avoided backing entire categories of businesses because of platform fees. 

Think about what we've missed - how many potential unicorns never got funded because investors knew the Apple Tax would break them?

The brief supports Epic Games, arguing that Apple's policies "stifle competition and innovation at its source."

Apple recently lowered fees to 27% for alternative payments, but imo this is more like a PR move than real reform.

YC said this is the first time in nearly two decades they can seriously consider investing in businesses that were previously impossible due to platform economics.

Two decades. That's an entire generation of mobile innovation constrained by a single company's pricing decisions.

The hearing is October 21st. Whatever happens, YC drawing a line in the sand matters for startups.

Every day thousands of people Google “convert .mov to .mp4” and land on sites like CloudConvert. Most of them use the free tier.

That sounds expensive…until you realise their entire model is just unit economics + funnels.

• Each conversion = compute + bandwidth.
• Free users cost them pennies in server time.
• A % hit file size limits, batch jobs, or API needs → upgrade to paid credits or API integrations.

The free site you and I see is really just top-of-funnel lead gen for an API company.

This is exactly the same playbook many AI startups are running.

Metered cost: CloudConvert pays per compute minute. AI wrappers pay per token.

-- Free tier as CAC: both give away usage until the limits pinch.
-- Upgrade path: heavier users pay for credits, subs, or API.
-- Enterprise layer: the real money comes from infra reliability, compliance, and sticky API contracts.

CloudConvert and Twilio turned “boring infra” into cash machines by nailing reliability + enterprise sales.

Most AI wrappers today are still thin layers over GPT - useful, but easily cloned without those moats.

If you’re building on AI, you’re really not just in the features game. You need to either:

1) Own a distribution wedge (like CloudConvert’s SEO), or
2) Climb the stack into infra + enterprise reliability.

Otherwise “free → paid funnel” just burns through your token bill.

Former OpenAI and DeepMind employees have launched 45+ startups worth $224 billion total.

That avg = $2.77 billion in enterprise value per founder.

Even excluding the top three companies, it's still $1.45 billion per founder.

𝗢𝗽𝗲𝗻𝗔𝗜 𝗔𝗹𝘂𝗺𝗻𝗶 𝗕𝗿𝗲𝗮𝗸𝗼𝘂𝘁:

  • Anthropic (Dario & Daniela Amodei): $15B valuation, $8B+ raised  
  • Safe Superintelligence (Ilya Sutskever, Daniel Gross, Daniel Levy): $32B valuation, $8B committed  
  • Thinking Machines Lab (Mira Murati): $10B valuation, $2B seed round  
  • Perplexity AI (Aravind Srinivas): $1B+ valuation, $100M+ raised  
  • Adept AI (David Luan): $1B+ valuation, $415M raised

𝗗𝗲𝗲𝗽𝗠𝗶𝗻𝗱 𝗔𝗹𝘂𝗺𝗻𝗶 𝗣𝗲𝗿𝗳𝗼𝗿𝗺𝗮𝗻𝗰𝗲:

  • Inflection AI (Mustafa Suleyman, Karén Simonyan): $4B valuation, $1.3B+ raised  
  • Mistral AI (Arthur Mensch): $6B valuation, $500M+ raised  
  • Harvey (Gabriel Pereyra): $500M+ valuation, $100M+ raised  
  • Daedalus (Jonas Schneider): $200M valuation, $34M raised

𝗠𝗲𝘁𝗮/𝗙𝗔𝗜𝗥 𝗔𝗹𝘂𝗺𝗻𝗶:

  • Cohere (Aidan Gomez, Nick Frosst): $2.2B valuation, $445M raised  EvolutionaryScale (Alexander Rives): \~$200M valuation, $142M raised  
  • Nabla (Alexandre LeBrun, Martin Raison): $800M valuation, $43M raised

𝗩𝗮𝗹𝘂𝗲 𝗖𝗿𝗲𝗮𝘁𝗶𝗼𝗻 𝗯𝘆 𝗦𝗼𝘂𝗿𝗰𝗲 (𝗲𝘅𝗰𝗹𝘂𝗱𝗶𝗻𝗴 𝘁𝗼𝗽 𝟯):

  • OpenAI alumni: $23.6B total valuation  
  • DeepMind alumni: $18.2B total valuation  
  • Meta/FAIR alumni: $3.4B total valuation  
  • Other sources: $1.3B total valuation

𝗙𝘂𝗻𝗱𝗶𝗻𝗴 𝗣𝗮𝘁𝘁𝗲𝗿𝗻𝘀:

  • DeepMind alumni raised: $3.89B total  
  • OpenAI alumni raised: $1.93B total  
  • Meta/FAIR alumni raised: $420M total

This super interesting data from Trace Cohen may explain Meta's $200M retention packages. It represents 0.0067% of their $1.5T market cap. Compare the ROI:

  • $100M = 0.13% of xAI's current valuation  
  • $100M = 0.31% of Safe Superintelligence
  • $100M = 1% of Thinking Machines Lab

Do these individuals represent the most concentrated value creation in startup history?

When Ilya Sutskever announces a company, investors commit billions before seeing a product roadmap.

The AI talent war is about the specific individuals who've proven they can architect systems that define technological paradigms.

🚨 New episode of Lobster Talks 🚨 

What if you could simulate how 1,000 investors will react to your pitch before you send it? 

That’s exactly how James He built Artificial Societies, raised capital, and got into YC. 

We go deep on: 
• Simulating human behavior with AI 
• Replacing market research and A/B tests 
• YC learnings & product-led growth 

🎙️ Watch the episode now → https://lnkd.in/ecMCm84Q 
🎧 Listen on the go: 
Spotify: https://lnkd.in/esphBCVZ 
Apple: https://apple.co/4cZ8RMm

Open article: The Lobster Talks Podcast by Lobster Capital by apple.co

The Lobster Talks Podcast by Lobster Capital

Taplio

Gabriel Jarrosson's Linkedin Analytics

What is Gabriel talking about?

's Best Posts (last 30 days)

Want to drive more opportunities from LinkedIn?

Taplio

.css-1jgreyo{background-image:linear-gradient(284deg, rgba(39,43,72,1) 0%, rgba(33,47,183,1) 100%);color:transparent;-webkit-background-clip:text;background-clip:text;}Gabriel Jarrosson's Linkedin Analytics

What is Gabriel talking about?

's Best Posts .css-qqfgvy{font-size:var(--chakra-fontSizes-md);}(last 30 days)

Want to drive more opportunities from LinkedIn?

Gabriel Jarrosson's Linkedin Analytics

's Best Posts (last 30 days)