Elias Mufarech

Great read from @jennywxiao: “the scarce and valuable layer becomes judgment: taste, prioritization, and context.” Imagine a world where humans are plugged into an ever-moving system, becoming nothing less than “judgment donors.” leonisnewsletter.substack.com/p/from-stopwat…

great interview

@BrandonGleklen i like your take Brandon, I think overall vibe diligence is a great thing if used well, it should push you to learn faster, digest 100% of a data room and remove the diligence “cold start problem” … but end of day you should always land in the main 2-3 questions that matter.

Google Research! Quantum getting much closer to reality. Willow chip ran a “Quantum Echoes” algorithm ~13,000× faster than the best supercomputer: real, verifiable quantum advantage, not just lab hype. New research says a future quantum box with <500k qubits could crack today’s Bitcoin/Ethereum-style crypto in minutes… ~20× fewer qubits than people thought, and Google is now targeting 2029 to move its own infrastructure to post‑quantum encryption. Source Google: Quantum Computers Need 20x Fewer Qubits to Crack Crypto bitbo.io/news/google-qu…

I think this clearly shows the direction of both OpenAI and Anthropic, consumer vs enterprise

On "kingmaking": No firm has articulated the logic of kingmaking more explicitly than Andreessen Horowitz, which in January 2026 announced $15 billion in fresh funds — more than the next two largest VC raises of 2025 combined. As Packy McCormick documented in a recent 16,000-word anatomy of the firm, a16z's culture doc holds that there is "no bigger investing sin" than backing the second-best company in a category. Once a16z identifies its candidate king, the classic move is to give it more money than it thought it needed. When Ben Horowitz met the Databricks founders and they asked for $200,000, he refused — and wrote a check for $10 million instead. As a16z partner Alex Danco put it, the firm's purpose is to function as a "legitimacy bank": conferring the credibility that lets a nascent startup sell to Fortune 500 buyers and hire senior talent it has no business competing for. That legitimacy now extends to a full in-house media operation — a "New Media" team led by Erik Torenberg that embeds with portfolio companies during product launches, runs timeline takeovers, and deploys storytellers to shape narratives in real time. When Flock Safety faced a PR crisis, a16z's team joined its Slack channels; when World Labs launched its Marble product, a16z's crew shot the announcement on a 3D LED volume stage. The capital is the headline. The machinery around it is the moat. vector.news/p/does-kingmak…

rationale? harvest guest content to automate CEO-level insights 🤣... honestly, might be a prime example of services as software

lot of rebuilding needed! PEs might need to build an FDE motion to help companies transition

agent-pilling is needed!

This debate about foundation models versus vertical solutions reminds me of the early internet. Back then, everyone thought they could just “go online” and sell products themselves.. but the reality was far more complex. To actually sell online you need: - A platform to upload products with correct descriptions - A payment processor - Automated invoicing - Packing and shipping with SLA-compliant visibility - Delivery notifications - Handling refunds, returns, and vendor coordination - Updating inventory and accounting Services firms choose specialized solutions over Claude because companies don’t just want tools, they want software that codifies workflows, ensuring predictable outcomes through autonomous systems. Agents can enhance efficiency, but they don’t replace the need for structured processes. True leverage comes from redesigning your company around these workflows.

Huge misunderstanding by everyone why companies buy software. Companies don’t want every employee doing every workflow from scratch on their own for every use case. At some point what you’re outsourcing is the ability to not have to think about the business process, and instead let the software provider think about it. Agents don’t change that, and probably if anything exhibit that dynamic even more.

B/c having a vendor w/ 1000’s of people who wake up every day literally only thinking about solving your problems is a good feeling and worth paying for and why the vertical software selloff is stupid because SMBs don’t fucking care about benchmarks they just want. it. to. work.

Financial statement re-shuffle across the P&L, balance sheet, and cash flow.. another example: Oracle Corporation is reportedly cutting 20–30K employees (~18%) to unlock ~$10B in annual cash flow, redirecting opex savings into debt-funded AI data center expansion.

@HarryStebbings quite a few nuances and caveats underneath this

Why a16z Should Back Every Single European Founder That Moves to the US: "We should back every single European founder who moves to the us, like we should just reflexively say yes. That's a combination of two things. 1. The raw level of talent that the great education system provides. 2. The move to the US indicates a willingness to seek risk and a greater level of achievement." @pmarca Love to hear your thoughts on this @t_blom @immad @dessaigne @Bandrew @joshbuckley @collinmathilde @maccaw @taavet

Strong insights for service companies building agents internally

@ImTheBigP simplifying hiring!

we all need to start moving to the right, re-thinking/engineering how work happens

good read about the future of org structures, with AI at the center of the entire organism, as a core intelligence mechanism

Rumored next-gen models (like Anthropic’s Mythos) could be 5–6x more expensive but dramatically more powerful. If so, AI starts to behave like a Veblen good, where higher price = higher demand. It might make sense that higher intelligence is almost always more expensive?

@chrija @Redpoint @loganbartlett Yep, but then you’re signing up to build and maintain a full communications product, with a low-margin operation and fairly average technical capacity

@eliasmufa @Redpoint @loganbartlett Completely agree. But most businesses are not fast-growing tech startups trying to become a generational company. Most companies are low-margin businesses where saving costs matters more.

This is a great report on the state of software and AI by @Redpoint - thank you, @loganbartlett! Where I disagree is the build vs. buy slide: 1) I'm not sure if it takes ~12 engineers to build/maintain a Slack clone for 1 customer. As AI keeps getting better at not only code gen but all software engineering tasks I think you'll be able to do it with a smaller team. Doesn't mean you should spend engineering time on it because I expect... 2) ... there will be agencies who specialize in this kind of work (e.g. build a Slack clone and sell customized versions of it). 3) ... there will be lots of cheap, (more or less) good enough Slack clones 4) ... there will be AI-native startups that rethink the category. All of these factors, I think, will contribute to pricing pressure for Slack and other traditional SaaS companies ... which they will only be able to defend against if they get a share of the agentic revenue enabled by their products.