Insights
Sharp takes on venture capital, AI, and cross-border dealmaking across X and LinkedIn.
I spent 18 months building an AI operating system for my own VC practice. Not a chatbot. Not a copilot. A full command center with its own agent, its own memory, its own budget.
The realization that changed everything: AI tools fail not because the models are bad, but because the interfaces are designed for the wrong century. We keep wrapping GPT-4 in chat windows and wondering why adoption stalls.
The interface IS the product. bakeryOS is my answer to the L1-to-L3 problem. And if this works for me, it works for every operator running a complex practice.
A family office in Riyadh told me something last week that I can't stop thinking about:
"We don't allocate to funds anymore. We build teams."
The in-house capital formation model is replacing the traditional LP-GP relationship for the largest family offices in the Gulf. They're hiring analysts, building proprietary deal pipelines, and cutting out the 2-and-20 entirely.
If you're a GP still sending quarterly letters and hoping for re-ups, the ground is shifting beneath you.
Unpopular opinion: The best investor relations happen at 2am in a WhatsApp group, not in a quarterly letter.
The GPs who understand this are winning the next vintage. The GPs who don't are writing 40-page updates that nobody reads.
Proximity beats formality. Every time.
After hosting 86 Tech Tuesday events in Dubai, here's what I've learned about community building in the Gulf:
1. Nobody cares about your brand until they've had coffee with you three times
2. The best conversations don't happen on stage -- they happen when someone misses their ride home
3. Consistency beats intensity. Show up every Tuesday for two years and the city starts to orbit around you
Build for serendipity. The dealflow follows.
Dubai doesn't just attract capital. Dubai attracts the people who move capital.
There's a difference. One is passive allocation. The other is a network effect that compounds every quarter as more family offices, GPs, and operators relocate here.
You don't come to Dubai for the tax rate. You come because this is where the next generation of cross-border deals gets structured.
LPs don't fund optimism. They fund endurance.
After sitting on both sides of the table for a decade, the pattern is unmistakable. The capital goes to the people who've already been tested -- who can describe exactly what happened when their thesis broke, and what they did next.
Your track record isn't your returns. It's your scar tissue.
I started creating content because I couldn't find the media I wanted to consume.
There's no shortage of VC content. There's a massive shortage of VC content that's honest about what actually happens between the term sheet and the wire transfer. The messy middle. The calls that don't go well. The deals that die in diligence.
That's what Venture Protocol covers. The protocol, not the press release.
The cross-border deal that looks simple on paper:
1. Founder in Tallinn
2. Holding company in DIFC
3. LP in Zurich
4. Customers in Riyadh
4 jurisdictions. 4 legal systems. 4 cultural protocols.
This is the new normal for Gulf-connected deals. And the GPs who can navigate this without a 6-month legal process are the ones getting the allocations.
People ask me why I build in public.
Because the best dealflow in 2026 comes from people who already know how you think. Every essay, every podcast episode, every thread is a filter. The founders who resonate reach out. The LPs who disagree don't waste our time.
Content isn't marketing. It's deal origination with a longer feedback loop.
New episode of Venture Protocol just dropped. This one is different.
Chris Haunschild has structured M&A deals across the US, Europe, and the Gulf for the last decade. We sat down in DIFC and talked about what actually happens when you try to close a cross-border deal -- the legal landmines, the cultural misreads, and why the best dealmakers are part lawyer, part anthropologist.
Link in comments.
Subscribe
Follow the conversation.
Real-time takes on venture capital, AI infrastructure, and building from Dubai. No algorithms needed -- just follow where the signal is.