New Chapter Releases Symbolic Verifier for LLM Evaluation
Chapter 3 on building a symbolic verifier for LLMs from scratch is now live: https://mng.bz/lZ5B . And with this, the first 176 pages of Build A Reasoning Model (From Scratch) are now available. This verifier is a useful method for evaluating LLMs. It's one of the more useful evaluation methods in our repertoire. In addition, it also doubles as an important component for reinforcement learning, which has become a cornerstone of LLM training in 2025. (But more on that in future chapters.)
Measurement Fuels Decisions, Not Just Story Confirmation
Is measurement a dirty word for marketers? I caught up with LinkedIn ’s Head of Ads Measurement, Jae O. last week at Advertising Week NY, and his response to that question really stuck with me: “When you’re trying to measure...
Ask First, Build Later: Avoid Wasting a Year
The fastest way to waste a year: Build something nobody asked for, spend six months perfecting it, launch it, hear crickets, then finally ask people what they actually want.
Early-Stage Agencies Cost More than They Deliver
The harsh truth about agencies at early-stage: You're paying them to learn. Just like any new hire, they won't hit their stride until month 3-4. If you're pre-PMF, that's a very expensive education for everyone involved. Consider contractors and freelancers first to de-risk.
Outrageous Optimism Is Essential for Founders' Success
Huge honor being featured by @bigthink this week. We talked about why “outrageous optimism” is a required skillset for founders. Startups are built by people who believe just enough to keep going when everything says stop. When the odds are 1000...
Balancing Hybrid Work with Essential Face‑to‑Face Interaction
Here’s a look at why face-to-face interaction is so important and how companies can make it happen effectively without losing the benefits of a hybrid workplace. https://t.co/wKCq6pHqhM via @bestcorpevents #teambuilding #futureofwork
2 Million Niche AI Models Await Your Business Ops
Most people still think GPT, Claude, Gemini (and maybe a few others) are the only AI models that exist. The real number? It's closer to 2 million! - Most of them are very niche. - They solve very specific issues. You only pay for the...
Creative Gifts and AI Tools Drive SaaS Q4 Success
This week’s Enterprise SaaS CEO/Founder Mastermind We covered five topics that every SaaS founder needs to think about, here’s what happened. 📦 Creative Marketing Ideas for Year-End We talked Q4 campaigns with $20-25K budgets. Members shared: sending bulky FedEx packages...
Vertical Focus Drives SaaS Success in Niche Markets
New SaaS CFO Podcast 🎙️ with Ari Bleemer, CEO of One Crew! ✅ Deep vertical focus wins ✅ Fundraising in “unsexy” markets ✅ Simple software contractors love Watch 👉 https://t.co/P4Zs83vpnP| #SaaS #AI #OneCrew https://t.co/7kZzckdZIX
Dagger’s Container Use Simplifies AI Agent Parallelism
Docker is overkill for running AI agents in parallel. @Dagger_io’s new Container Use is a much simpler, agent-first alternative: → No Dockerfile needed → Git-native isolation → Full logs & real‑time visibility + easy one-liner commands to do it all! 🙌 Let’s compare 🧵↓ https://t.co/etVd1uHheZ
No‑code SaaS Creates Custom AI B‑roll in Hours
I turned this into a new SaaS 👀 Took 2 hours using @lovable_dev and @supabase (their new cloud integration is literally insane). Now it's much easier: - Upload your images. - Create a character. - Create AI B-roll clips. I didn't write (or read) a single...
OpenAI Slashes Microsoft Revenue Share, Gains $50B
Major shift at OpenAI: revenue share with Microsoft & partners to fall from 20% → 8% by end of decade, potentially netting @OpenAI an extra $50B. Big move on margins, and implying pressure on terms and structure. 🔗 https://t.co/kjyCLj1C6B #AI...
Even LLMs Can't Answer Simple Emoji Queries
Some questions shouldn't be asked of LLMs. :-)
Top AI Minds Ranked: Who Defines Our Future
All the best in AI are on this list: https://t.co/2dDZtIT7Gf
Your Quick Guide to Understanding Salesforce
If you ever wanted to know what Salesforce is! 🤘 https://t.co/k8UNyYc8HD
Prioritize Delighted Customers over Hype Growth Curves
Dear Hyper-growth founders, Legendary companies are built on the backs of their delighted customers, not on the backs of their buzz/curves/demos. Don't fall in love with your hockey stick curves...fall in love with your customers. Love, Brian.
Transparent Prep and Dialogue: CEO Board Meeting Essentials
Board meeting tips for CEO's of private companies: 1. Be transparent -- it builds trust. Board members can tell if you are spinning them. Save the sales pitch for the sales prospect. 2. Prep time -- Either send...
Add AI to Any Site with a Tiny Widget
[tiny project] ai-share-widget the easiest way to add* ai capability to a site with a front-end widget where you can customize the prompt and allow the user to pass that into their AI tool of choice github (102 lines of code) in...
AI Agents Compress Software Development Cycles to Hours
So many of our traditional workflows and processes for building software will be rebuilt because of AI agents. For instance, when building new features using an AI agent, it’s a wild because you basically will evolve the product’s requirements and spec...
Every Startup Launch Instantly Meets a Twin Competitor
Every new startup launch seems to have a twin. Either a competitor already exists or one gets announced within days. Add every incumbent quietly building the same thing, and it’s no wonder products feel like déjà vu. Been on all...
Video AI Faces Its Own Turing Test
The Turing Test for video … 😅
Free Access for 20: New AppSumo Launch Tomorrow
Launching on AppSumo tomorrow. Guess what it is - I'll give it FREE to 20 people.
Walmart Explores ChatGPT-Powered Shopping Experience
Walmart to allow customers to shop using ChatGPT? https://t.co/GcOFDzhy0f #shopping #onlineshopping #wallmart #ChatGPT @PawlowskiMario @chidambara09 @Ym78200 @CurieuxExplorer @efipm @bigfundu @sayedflah @Ronald_vanLoon @cyngn @belindabeibi @odisseiaalfa @DigitalColmer @MyCompanionsAI @KirkDBorne @patricegorissen @jeffkagan @EdwardKens50830 @enilev @insom_ai333 @ChrisCCrowley @sallyeaves @gurmeet_judge @VairagyaSadhana @andresvilarino @amomsimpression @amit6060 @0xAmol @OfficialDabier
Product-Led Growth Trumps Sales with Frictionless Onboarding
Why Product-Led Growth Beats Traditional Sales!! In this video, we explore how self-service acquisition, frictionless user onboarding, and natural value expansion create a seamless experience for users. Learn the secrets behind viral loop architecture and how the right product design...
Mac Mini Excels at Inference; DGX Spark Needed for Fine‑tuning
Saw that DGX Spark vs Mac Mini M4 Pro benchmark plot making the rounds (looks like it came from @lmsysorg). Thought I’d share a few notes as someone who actually uses a Mac Mini M4 Pro and has been tempted...
Codex CLI Exit Saves Sessions, Huge QOL Boost
whoever added this to the codex cli when you exit is a lifesaver cannot tell you how many times I've mindlessly ctrl+c trying to copy paste only to break a session then i gotta dig up the latest chat. such a simple...
AI Art Hits Sameness, Lacks Emotional Storytelling
AI is amazing but it isn't good enough yet. @Kantrowitz says, below, AI has a sameness problem. The problem is if you hang out here on X all day long we see the sameness. Look at my AI Artist's feed. If...
OpenAI Targets $1 Trillion From $13 Billion in Five Years
"OpenAI has five years to turn $13 billion into $1 trillion." Do you think they can do it? https://t.co/py7EgsSvnP
Acquired.com Launches New Seller Insights and Analytics
New seller insights and analytics coming soon to @acquiredotcom! https://t.co/V2mQKaoOFj
One-Day Prototype Loop Redefines Product Development
I just recorded a pod with @thisisgrantlee (CEO of @MeetGamma), and he described a way his product team operates that I think is a glimpse into the future of how product teams will operate: 1. Team member has an idea in...
NanoChat Code Sparks Fresh Look at LLM Theory
Karpathy’s nanochat codebase motivated me to revise llm theory
Anthropic Releases Comprehensive Claude Code Blueprint for Developers
Anthropic just published a killer blueprint for Claude Code users! 🔥 → GitHub CLI workflows → .md prompt tuning → Custom slash commands → Headless + multi-agent agents Dense, yes... but pure gold if you're building with LLMs. Download link in 🧵↓ https://t.co/kCbF9amMFW
Let Vendors Handle AI Agent Deployment, Not You
You probably need a team of Forward Deployed Engineers. Even if you don't think you do. We've deployed 11 AI Agents so far at SaaStr, and we're adding our 12th, Agentforce. More on that soon! One common thread: the vendor...
Launch of Readsail: Curated AI Learning Platform
If your company wants to help employees stay up to date with AI, I've teamed up with @natolambert and others to launch @readsail. It's essentially a platform that makes it easy to manage ongoing AI learning and brings you a...
From Moonshot to Millions: Veo’s Video Revolution
In 2018, video generation was a moonshot project by the Google Brain team. Today, hundreds of millions of videos have been generated with Veo. 🤯 In this episode of Release Notes, @doomie and @OfficialLoganK reflect on Veo's journey from research to...
AI Contests Accelerate Product Fixes and Empower Creators
Since I just judged another, similar, contest, these are smart for companies to do. First, they are a forcing function to get AI companies to ship and fix things, but they get feedback from real users and get lots of...
Join Fizzy Private Beta: Early Access Before Public Launch
Aiming to begin sending out a handful of Fizzy private beta invites as soon as tomorrow. If you want to get on the list, quickly sign up here: https://t.co/6eFNse2WPf Also... Promise that if you sign up on the list you *will* get an...
AI Autonomously Resolves IT Outages, Easing Human Overload
AI is stepping in as an autonomous operator for overloaded humans. ➡️ AI agent detected VPN outage ➡️ Authenticated into New Relic ➡️ Re-established expired integration ➡️ Streamed telemetry data ➡️ Identified SW update as root cause ➡️ Suggested a rollback IT lead reviewed...
Watch Dreamforce 2025 Keynote Now Online
The Dreamforce 2025 keynote is now online! https://t.co/q9FDsk9mBP.
AI Enhances Human Creativity and Boosts Event Value
Whirlwind trip to #Dreamforce25 this week, amazing to get time with friends, hang at 6sense 's #Club6 , talk agentic marketing with CMOs at the Qualified breakfast, then hang with the one and only Brent Adamson at his book signing....
AI Will Boost Selling Time to 80% by 2026
Most GTM teams haven’t even entered Phase 1. That’s the punchline of this new piece I wrote regarding my hypothesis on how AI will reshape field of go-to-market over the next decade. In the work, I propose four sequential phases...
Nano Banana Expands to Slides, NotebookLM, Gemini Tools
Several of our launches this week have been absolutely 🍌… Here are the new places you can find Nano Banana: — Edit images using custom prompts directly in Slides @googleworkspace — @NotebookLM’s video overviews got a huge upgrade with 6 new...
Dreamina AI Surpasses Nano‑Banana, Offers Multi‑Media Creation
I'm playing with @dreamina_ai (used it on my post earlier) and it's quite special for creating a wide variety of things, videos, photos, posters. It's from ByteDance and is beating the others, lets you upload six images and mix them...
Brand, Not Hacks, Is 2025's True Marketing Moat
People talk about endless growth hacks, but is the REAL marketing moat in 2025 Brand? 🎥 Search "Leveling Up with Eric Siu" on YouTube for more contents like this https://t.co/u5dOhYSmqT
Underfund Sales & Marketing Limits SaaS Growth Past $5M
The biggest mistake SaaS founders make is spending too little on sales & marketing. This works until around $3M–$5M ARR, not more. Here’s why👇