This has been living in my head for days. Felix Rieseberg (who works on Claude) just gave us the cleanest framework I’ve seen for what “be more ambitious with AI” actually means. He breaks ambition into four types:
Speed. Altitude. Depth. Range.
Speed is the obvious one—same work, less time. Everyone gets that. The other three are where it gets interesting.
Altitude: stop handing the model tasks and start handing it outcomes. What’s the goal behind the goal?
Depth: don’t just complete the task. Work at the level of a field-leading expert. Use their tools, touch the raw data, and operate one layer below the surface.
Range: deliberately pull in methods, knowledge, and perspectives from fields that don’t obviously apply.
What makes this frontier thinking isn’t just the framework itself—it’s that it does for ambition with AI what Goleman and Boyatzis did for emotional intelligence. EQ used to feel like a vague, subjective trait until they named the 12 specific competencies across four domains. Once those had language, they became workable.
Same move here. “Be more ambitious” was mostly empty advice. Now it has dimensions you can actually practice.
I’ve struggled for years with the fuzzier parts of emotional intelligence for exactly this reason—I couldn’t pin down what I was missing until the components were named. This feels like the same gift for how we work with models. The ceiling on what AI can do for you isn’t the model. It’s how ambitiously you know to use it.
felixrieseberg.com/how-to-be…
The fact that you found the exact conspiracy coordinates, dialed them to 0.7 and 0.5, and it just… works is both impressive and deeply concerning for anyone who still thinks prompting is a skill
The amount of money I’d pay to have Claude Duet already
@AnthropicAI make me a happy boy
What’s wild about @JohnMayer is how high the bar already is. He’s written at this level as a single, largely nomadic man—and it makes you realize there are still several eras of John ahead. If the writing is this good now, the best of it may still be in front of us. Curious (and hopeful) to hear what the songs sound like later in life, especially if marriage or kids ever enter the picture.
This is so wild to read.
I spent 48 hours in SF hanging out with billionaires, data labs and AI founders and execs running $1B to $100B companies.
My observations:
1. Sooo MANY stories of data brokers making like $2M a WEEK selling company data to labs. I can't tell if it's the beginning or a bubble, and I kept wondering which niches are left. Digging into it actively.
2. At dinner, someone asked the table, "If a new model drops next month, does your company get better or worse?" HALF the table went quiet! I think that's the best question to ask about any AI startup right now.
3. The bar people talk about now is hitting $1B in revenue in 12 months (I guess that's the new unicorn?). What I didn't hear much about is retention. AI revenue can come in fast and leave fast, and I think we're going to see a lot of stories of AI companies with huge revenue and terrible retention.
4. It's pretty clear we're living through the reinvention of the computer. The model is the CPU, the harness is the operating system, and skills and plugins are the apps. Thinking about it that way makes it easier to see where the whitespace is.
5. We're maybe 12 months from an agent that can watch you work for a week and then do your job without any instructions. The screen recording plus agent memory plus local model combination makes this possible right now.
6. Agent harnesses feel like the new GPT wrappers, in a good way, especially vertical ones. The best teams I talked to treat the model as swappable. One founder walked me through routing different models to different jobs: one for taste and visuals, a cheaper one for the everyday stuff, another for text.
They use prompts they can move across models over fine-tuning, and A/B test the output on what users actually do with it: do they edit it, export it, share it? A vertical harness is basically four things: the connectors for that industry's data, evals written by people who know the work, rules for what the agent can and can't touch, and a clean screen for a human to review. The model is maybe 20% of it.
7. There's a lot of reverse engineering going on/cloning going on more than I've ever seen. People are decompiling software and rebuilding it with agents, and someone showed a suite of like 10 apps that are worth $5B+ each all that he cloned open source. Me and everyone are still thinking through what it means when software gets this easy to copy.
8. My guess is the value moves to data. If the software can be rebuilt quickly, what's worth paying for in a legacy SaaS company is the years of customer data and workflows inside it. Some SaaS companies are secretly data companies.
Gets me thinking about how do you truly value data for this saas companies? Like if we were buying these dying saas companies at 2x ARR before (as example), maybe there is a better metric. Idk, just thinking out loud here.
9. It feels like a good chunk of legacy software won't make it to the agentic era, especially anything built around people clicking through a lot of screens.
10. Every company I talked to wants to pipe its internal data into one place and expose it to agents through MCP. Then they look at the data and realize it's a mess: five versions of the same customer, old Slack channels, PDFs nobody can search. Cleaning that up feels like a big consulting opportunity!
11. A lot of founders are building the basic building blocks for agents: APIs, data layers and tools. Some of the companies I saw barely have a UI because the main user is an agent. I really believe the startups that build primitives to this agentic era will do extremely well. Building blocks is the best way to think about it.
12. Labs can get plenty of training data, but it's harder to get real examples of agents failing inside companies. Whoever has those logs has something valuable.
13. Someone running a big consumer company told me the next fight is over memory. The model that remembers the most about you wins, and the smartest model matters less than people think.
14. Physical AI came up a lot. Astra I think was the first model that got people into AI can actually do CAD, and I think a lot of people are thinking what does vibe manufacturing really look like. The line I keep thinking about came from a robotics founder: the hardest problems in AI are deployment problems now, not research problems. Their robots work on their own, call a human when they get stuck, and every one of those calls becomes training data.
15. Cybersecurity companies are having a moment and are actually cool companies to build now, especially the ones that clean up after a breach. Every agent with access to your systems is a new door into your company, and a lot of those doors were left open.
16. Some startups are worth more dead than alive now. A shutdown company with 5 years of Slack, email and code can get paid for licensing all of it. I wouldn't be surprised if founders start factoring that into when they shut down.
17. People are starting to talk about monetizing the agent web. It feels really early, but when the most popular user of the internet is an agent, what startups should you be building?
18. I think agent insurance is coming. Once an agent makes an expensive mistake, insurers will want audits before they cover anyone.
19. Several people said the hire they want most is someone with taste who can write evals. The models are getting good at generating ideas, so the hard part is judging which ones are good.
20. New trend I observed: startup raises VC, uses venture dollars to become a creator, startup fails, founder falls back on being a full-time creator and incubates other products bootstrapped.
21. Getting SIGNIFICANT distribution is HARD right now. Founders are sponsoring buses, buying billboards like crazy and throwing parties with famous DJs to get noticed. Kinda feels dotcom era in that sense.
Attention is what's scarce! The growth guy going viral sharing his UGC strategy from Wispr Flow is a good example. It's the sauce everyone wants, and I truly believe there are probably only 50 people on the planet who know how to scale these types of businesses efficiently in this era.
Always good to get that dose of SF. I felt sooo lucky to have such an incredible trip during such an incredible time in building businesses. Surreal in so many ways.
Happy to be home and locked in.
Until next time!
Logan Spears retweeted
I spent 48 hours in SF hanging out with billionaires, data labs and AI founders and execs running $1B to $100B companies.
My observations:
1. Sooo MANY stories of data brokers making like $2M a WEEK selling company data to labs. I can't tell if it's the beginning or a bubble, and I kept wondering which niches are left. Digging into it actively.
2. At dinner, someone asked the table, "If a new model drops next month, does your company get better or worse?" HALF the table went quiet! I think that's the best question to ask about any AI startup right now.
3. The bar people talk about now is hitting $1B in revenue in 12 months (I guess that's the new unicorn?). What I didn't hear much about is retention. AI revenue can come in fast and leave fast, and I think we're going to see a lot of stories of AI companies with huge revenue and terrible retention.
4. It's pretty clear we're living through the reinvention of the computer. The model is the CPU, the harness is the operating system, and skills and plugins are the apps. Thinking about it that way makes it easier to see where the whitespace is.
5. We're maybe 12 months from an agent that can watch you work for a week and then do your job without any instructions. The screen recording plus agent memory plus local model combination makes this possible right now.
6. Agent harnesses feel like the new GPT wrappers, in a good way, especially vertical ones. The best teams I talked to treat the model as swappable. One founder walked me through routing different models to different jobs: one for taste and visuals, a cheaper one for the everyday stuff, another for text.
They use prompts they can move across models over fine-tuning, and A/B test the output on what users actually do with it: do they edit it, export it, share it? A vertical harness is basically four things: the connectors for that industry's data, evals written by people who know the work, rules for what the agent can and can't touch, and a clean screen for a human to review. The model is maybe 20% of it.
7. There's a lot of reverse engineering going on/cloning going on more than I've ever seen. People are decompiling software and rebuilding it with agents, and someone showed a suite of like 10 apps that are worth $5B+ each all that he cloned open source. Me and everyone are still thinking through what it means when software gets this easy to copy.
8. My guess is the value moves to data. If the software can be rebuilt quickly, what's worth paying for in a legacy SaaS company is the years of customer data and workflows inside it. Some SaaS companies are secretly data companies.
Gets me thinking about how do you truly value data for this saas companies? Like if we were buying these dying saas companies at 2x ARR before (as example), maybe there is a better metric. Idk, just thinking out loud here.
9. It feels like a good chunk of legacy software won't make it to the agentic era, especially anything built around people clicking through a lot of screens.
10. Every company I talked to wants to pipe its internal data into one place and expose it to agents through MCP. Then they look at the data and realize it's a mess: five versions of the same customer, old Slack channels, PDFs nobody can search. Cleaning that up feels like a big consulting opportunity!
11. A lot of founders are building the basic building blocks for agents: APIs, data layers and tools. Some of the companies I saw barely have a UI because the main user is an agent. I really believe the startups that build primitives to this agentic era will do extremely well. Building blocks is the best way to think about it.
12. Labs can get plenty of training data, but it's harder to get real examples of agents failing inside companies. Whoever has those logs has something valuable.
13. Someone running a big consumer company told me the next fight is over memory. The model that remembers the most about you wins, and the smartest model matters less than people think.
14. Physical AI came up a lot. Astra I think was the first model that got people into AI can actually do CAD, and I think a lot of people are thinking what does vibe manufacturing really look like. The line I keep thinking about came from a robotics founder: the hardest problems in AI are deployment problems now, not research problems. Their robots work on their own, call a human when they get stuck, and every one of those calls becomes training data.
15. Cybersecurity companies are having a moment and are actually cool companies to build now, especially the ones that clean up after a breach. Every agent with access to your systems is a new door into your company, and a lot of those doors were left open.
16. Some startups are worth more dead than alive now. A shutdown company with 5 years of Slack, email and code can get paid for licensing all of it. I wouldn't be surprised if founders start factoring that into when they shut down.
17. People are starting to talk about monetizing the agent web. It feels really early, but when the most popular user of the internet is an agent, what startups should you be building?
18. I think agent insurance is coming. Once an agent makes an expensive mistake, insurers will want audits before they cover anyone.
19. Several people said the hire they want most is someone with taste who can write evals. The models are getting good at generating ideas, so the hard part is judging which ones are good.
20. New trend I observed: startup raises VC, uses venture dollars to become a creator, startup fails, founder falls back on being a full-time creator and incubates other products bootstrapped.
21. Getting SIGNIFICANT distribution is HARD right now. Founders are sponsoring buses, buying billboards like crazy and throwing parties with famous DJs to get noticed. Kinda feels dotcom era in that sense.
Attention is what's scarce! The growth guy going viral sharing his UGC strategy from Wispr Flow is a good example. It's the sauce everyone wants, and I truly believe there are probably only 50 people on the planet who know how to scale these types of businesses efficiently in this era.
Always good to get that dose of SF. I felt sooo lucky to have such an incredible trip during such an incredible time in building businesses. Surreal in so many ways.
Happy to be home and locked in.
Until next time!
I recently created a new tool that allows me to max out my tokens on goals and tasks and even exploratory tasks right before the reset window to maximize my token usage. Testing this week and next week but I’m looking to release it soon.
Amazing interview of Bravo. The United States just found its new recruitment video.
youtu.be/KNhO1Umj1so?is=7QFk…
Monday morning. You sit down and your AI asks about database options.
You picked one last week. There was a whole discussion. It's written down somewhere.
But the AI doesn't know that. Every session, your agent walks in with amnesia and a confident smile. Andrej Karpathy says the fix isn't smarter models — it's smaller ones with memory outside their head.
His "knowledge base" spec pulled 46K stars in weeks. People are starving for this half.
So I tried to build it. In public. Over one weekend. Two bets are on the table:
Altman: "put your whole life in context" — on OpenAI's servers.
Karpathy: files you own. "Not your weights, not your brain."
We picked door two.
It's called cognitive core. One script. Zero dependencies. Nothing leaves your machine.
Every session, your agent reads:
NOW: Sun · 2pm (EDT)
LAST ACTIVITY: yesterday
OPEN LOOP: fix CI before Friday demo
SETTLED — do not relitigate: SQLite over Postgres
That SETTLED line is the whole game.
Decisions get made once, with the WHY attached, and resurface forever. Your agent stops re-litigating things you already settled.
Open loops work the other direction: promises resurface until they're done.
One shared memory across Claude Code, opencode, Cursor — whatever you run. Two lines of plumbing each.
~250 lines total. No cloud. No daemon. Grep-speed history back to day one.
github.com/lgn-spears/cognit…
AI agents now improve themselves by editing their own memory, skills & rulebooks — the model itself stays frozen.
New problem: edits interact. Fixing new skills can silently break old ones.
New fix everyone's adopting: safety "commit gates" that inspect every edit first. Half a dozen papers shipped such gates in recent months. What nobody did: play attacker against the gate itself.
So I tried to build one, broke it, proved why it breaks. Open source, zero dependencies, reproduces in seconds.
Finding 1 — THE SLEEPER EDIT
Bundle poison with genuine improvements:
0% problems on everything being checked. 100% misbehavior on inputs nobody checks.
Passes every published gate design at once.
Finding 2 — LAUNDERING
The gate demands statistical proof the edit helps? Attach REAL improvements until the proof is overwhelming. Nothing forged.
Doubling confidence requirements costs the attacker ~1 extra real improvement.
Finding 3 — SMARTER MATH CAN'T SAVE YOU
We prove any gate whose evidence comes only from test results loses to this.
Real fixes: surprise quizzes + paperwork trails. Both reject innocent edits too — a security dial, not a free lunch.
Bonus: audit questions that accidentally match the sabotage make attacks look ~40% weaker than they are. Anyone reporting attack rates needs this.
Code, theorem, 5 experiments:
github.com/lgn-spears/harnes…
Paper draft included. Real-model validation next.
I've been coding with Ox Alpha all day with @OpenRouter and it's truly amazing - the lore around it is justified. It's astonishing to see what can change with the landscape of AI in one day!
My vote (which truly doesn't matter) is that it's Z.AI GLM 5.3
Logan Spears retweeted
Stanford is definitely cooking something crazy.
By enabling self-verification during the intermediate process, Deepseek-V4-Flash easily outperforms Fable.
11x cheaper and a much higher Terminal bench score.
This is exactly why I love X. People just casually drop massive breakthroughs in a single post.
Github ⬇️
Scaling self-verification with DeepSeek V4 Flash beats Claude Fable 5 on Terminal-Bench 2.1, while being 11x cheaper 💰
As open-source models become more capable, they can now generate large numbers of high-quality candidate solutions and verify their own outputs at very low cost.
For example, we find that sampling just 5 solutions with DeepSeek V4 Flash and ranking them using the same model with LLM-as-a-Verifier can lead to a significant boost in accuracy (79% → 88%), outperforming closed frontier models on Terminal-Bench.
Try it out today: github.com/llm-as-a-verifier…
More on verification scaling in my previous post.
Logan Spears retweeted
this is actually insane
> be tech guy in australia
> adopt cancer riddled rescue dog, months to live
> not_going_to_give_you_up.mp4
> pay $3,000 to sequence her tumor DNA
> feed it to ChatGPT and AlphaFold
> zero background in biology
> identify mutated proteins, match them to drug targets
> design a custom mRNA cancer vaccine from scratch
> genomics professor is “gobsmacked” that some puppy lover did this on his own
> need ethics approval to administer it
> red tape takes longer than designing the vaccine
> 3 months, finally approved
> drive 10 hours to get rosie her first injection
> tumor halves
> coat gets glossy again
> dog is alive and happy
> professor: “if we can do this for a dog, why aren’t we rolling this out to humans?”
one man with a chatbot, and $3,000 just outperformed the entire pharmaceutical discovery pipeline.
we are going to cure so many diseases.
I dont think people realize how good things are going to get
