Free AI Content Engine📝, AI Costs Keep Rising 💸, How to Kill Churn📉
Building, investing, or staying sharp on what’s moving. This is your weekly edge.
Insights, funding news, and the VC funds deploying capital right now. Just what matters.
Let’s get into it 👇
The founders winning in 2026 are everywhere. LinkedIn, newsletters, podcasts, investor DMs.
They run on a system. And this week, we are giving that system away for free:
Together with Notion, we built a free AI Creator Kit: one Notion workspace with AI agents, templates, a creator CRM, and partner credits worth thousands. One weekend to set up. Compounding starts the week after.
What is inside:
▫️ Memory-persistent AI content agent that holds your brand voice
▫️ AI-powered social media calendar built in Notion
▫️ Creator CRM for partnerships and outreach
▫️ Slack, email, and meeting notes syncing automatically
▫️ 3 months of Notion Business free ($6K value), plus CapCut, Lovable, and Secret credits
In-Depth Insights 🔍
Elon Musk’s 2007 Interview Predicted the Next 18 Years ⏳
A forgotten conversation mapped product pricing, launch order, and execution timelines years before Tesla and SpaceX proved the thesis. The value is not prediction but the discipline of thinking through a long sequence before shipping the first product.Building in Public Is Not What It Used To Be 🧠
The strongest public content explains decisions, tradeoffs, and reasoning instead of exposing operational details anyone can copy. Trust compounds through judgment while metrics, playbooks, and channels often create more downside than advantage.
The Real Reason AI Costs Keep Rising 💸
Lower token prices made complex agent workflows practical, pushing overall spending higher instead of lower. The new constraint is managing usage efficiently as finance teams begin tracking tokens like any other operating cost.
What Actually Moves the Needle with Claude ⚙️
A practical framework focuses on repeatable workflows, structured context, and verification instead of one off prompting. The guide combines templates, habits, and setup routines designed to improve output quality over time.
The Claude Guide Every Founder Should Run Before Fundraising 🧾
Many investors now use automated screening to review pitch decks before they reach a partner’s inbox. Running the same checks yourself helps catch weak claims, missing context, and inconsistencies before submission. [Chris Tottman]
What Is Talent and Talent Engineering 💼
New hiring roles are emerging around finding, evaluating, and enabling exceptional people inside fast moving companies. The article explains what these positions actually cover beyond the labels appearing across job boards. [Rich]
How to Kill Churn: Everything I Learned in 10 Years 📉
Lessons from scaling VEED highlight how retention strategies change across self serve, enterprise, and API customers. Understanding why each segment leaves leads to better prioritization than applying one playbook to every user. [Sabba Keynejad]
Idea Generators 💡🌐
Long running conversations between experienced founders became a steady source of new business ideas and market perspectives. Consistent intellectual sparring often produces stronger opportunities than waiting for isolated moments of inspiration. [Reid Hoffman]
Social Media Gems 💎
World Cup 2026 Bracket Design 🏆
A redesigned tournament bracket turns a familiar format into something people want to explore. The strongest visual systems improve understanding by changing the experience, not adding complexity.
Developers Developers Developers 👨💻
The race is no longer just about building better models but winning the developers using them. Control the builder ecosystem, and you shape the products, distribution, and long term platform advantage.Chamath Names Himself CEO 💰
8090 Labs raised $135M with backing from major investors and brought Chamath back into a full time operating role. The company is focused on AI generated enterprise software with auditability built for regulated industries.
Trump’s Semiconductor Heavy Portfolio 📊
The latest disclosed allocation shows a clear preference for semiconductor companies across the portfolio. Large positions in NVIDIA, Qualcomm, Micron, and Intel point to conviction in chip infrastructure. [Charles-Henry Monchau, CFA, CMT, CAIA]
Trending News ⚡
Firmus Locks In Massive Nvidia Compute Supply 🤝
Firmus secured access to 170,000 GPUs for delivery between 2027 and early 2028 through a strategic Nvidia partnership. The agreement pairs infrastructure purchases with cloud services and projects up to $30 billion in revenue over six years. [Reuters]
Samsung Commits $90 Billion to Chip Expansion 💰
The investment spans display panels, memory packaging, batteries, and semiconductor materials across multiple Samsung units. The plan deepens domestic manufacturing capacity while supporting future demand for advanced computing infrastructure. [Reuters]Creative Design Remains a Tough Problem for AI 🎨
OpenAI’s Codex lead says evaluating design quality lacks the clear feedback loops that make coding easier to train. Figma’s CEO argues models often converge toward familiar patterns, leaving originality as a human advantage. [Business Insider]
Meta Says AI Agents Need More Time 🤖
Mark Zuckerberg told employees development has progressed more slowly than internal expectations. Meta also shifted its employee activity tracking program toward an opt in approach after internal criticism. [Business Insider]
Anthropic Explores Samsung Chip Partnership 🤝
The companies are discussing a custom processor without committing to a final product or deployment plan. Anthropic continues to spread compute demand across multiple hardware partners instead of relying on one supplier. [TechCrunch]
Micron Rides the AI Memory Wave 📈
Strong demand for high performance memory pushed the company to record revenue and profit growth. Long term supply agreements with major customers reinforce memory as a critical layer of AI infrastructure. [TechCrunch]
SpaceX Prototype Report Sparks Hardware Speculation 📱
Reports claim investors saw an early AI handset concept built around xAI software integration. Elon Musk rejected the story while acknowledging any future hardware remains at an early stage. [TechCrunch]
Base44 Builds Its Own Model for Better Economics 🧩
The company trained Base1 using millions of real platform interactions instead of relying entirely on external models. Owning the stack gives greater control over speed, operating costs, and product optimization as revenue scales. [TechCrunch]
New Funds 💰
PranaVentures launched a €100M seed fund to invest in early-stage startups across Europe.
Stratos Ventures raised over $50M for its debut fund focused on backing early-stage technology startups.
InvestEco closed Sustainable Food Fund IV at C$106M to invest in sustainable food and agri-tech companies.
Tapestry VC launched $80M Fund III to continue investing in enterprise infrastructure startups.
Harpoon Ventures closed Fund IV at $155M to back frontier and defense technology startups.
Ruya Ventures closed its inaugural $50M pre-seed fund to support founders across the MENA region.
Magnify Ventures closed Fund II at $46.6M to invest in deep tech and life sciences startups.
That’s a wrap for this week.
If you want to stay ahead of the curve, make sure you're subscribed. No fluff, just real startup insights delivered straight to your inbox.
See you next time 🚀
I liked the part with the evaluation of missing context, it is indeed needed. Though, evaluating the AI output is the tiny part of the whole process.
Hey wanna give eachother a little mutual boost and like eachothers recent posts ?