MFH

| former aviation finance | masters game theory | jedi

[MFH @fardeeze]

@mann_idan @togethercompute @FireworksAI_HQ https://t.co/3bjc3fNRVD

[Unnati @Unnati_builds24]

Whenever I feel demotivated i rewatch this...

Elon Musk in a living room in 2008, 12 days from bankruptcy, pitching a $109,000 car to people...

[jack @jack]

there is no moat

[MFH @fardeeze]

5. Domain Obsession
You can't relentlessly execute if you aren't obsessed with the problem.

Mastery of the mundane: They don't just know the sexy parts of their business; they know the boring, granular details. They know the exact unit economics, the weird regulatory loopholes, and the mundane daily workflows of their target customers.

Unprompted depth: When asked a surface-level question, they can easily drop five layers deep into the nuance of the market without missing a beat.

The Ultimate "Silent" Signal
The most definitive proof often happens after the meeting ends. The follow-up from a "get it done" founder is fast, immaculately organized, and addresses every single open thread from the conversation, usually with additional data attached. 🧵 6/6

[MFH @fardeeze]

Here are the specific signs and attributes evaluators look for to determine if a founder has that relentless execution gene. 🧵 1/6

[MFH @fardeeze]

4. An Internal Locus of Control
This is a psychological term meaning the person believes they are responsible for their own success or failure, rather than attributing it to outside forces.

Owning failures: When discussing a past startup failure or a major mistake, they say, "I misjudged the market," or "I hired the wrong person." They never say, "The investors didn't get it," or "The economy ruined our launch."

Refusal to play the victim: They don't complain about how hard it is to raise money, how unfair a competitor's advantage is, or how tough the regulatory environment is. They view obstacles as puzzles, not excuses.
🧵 5/6

[MFH @fardeeze]

@AnneliesGamble Maybe inference needs a financial layer too? https://t.co/dmotKhcjwp

[MFH @fardeeze]

Dubai’s first real stress test

Flying colors thus far

[MFH @fardeeze]

Breaking news:

France have surrendered!

[MFH @fardeeze]

Excellent. Super plausible and scary

[Citrini @citrini]

JUNE 2028.

The S&P is down 38% from its highs. Unemployment just printed 10.2%. Private credit is unraveling. Prime mortgages are cracking. AI didn’t disappoint. It exceeded every expectation.

What happened?​​​​​​​​​​​​​​​​

https://t.co/JzzwCrbJgS

[MFH @fardeeze]

@lordOfAFew This is why you’ll need oracles like @Exostreamai

[Exostream @Exostreamai]

1/ LLM tokens are commodities. There are 29 models across 6 providers, prices change constantly, and there’s no structured feed tracking any of it. So we built one. https://t.co/ch4txI0V9q — the pricing oracle for LLM inference.

[Andrej Karpathy @karpathy]

A few random notes from claude coding quite a bit last few weeks.

Coding workflow. Given the latest lift in LLM coding capability, like many others I rapidly went from about 80% manual+autocomplete coding and 20% agents in November to 80% agent coding and 20% edits+touchups in December. i.e. I really am mostly programming in English now, a bit sheepishly telling the LLM what code to write... in words. It hurts the ego a bit but the power to operate over software in large "code actions" is just too net useful, especially once you adapt to it, configure it, learn to use it, and wrap your head around what it can and cannot do. This is easily the biggest change to my basic coding workflow in ~2 decades of programming and it happened over the course of a few weeks. I'd expect something similar to be happening to well into double digit percent of engineers out there, while the awareness of it in the general population feels well into low single digit percent.

IDEs/agent swarms/fallability. Both the "no need for IDE anymore" hype and the "agent swarm" hype is imo too much for right now. The models definitely still make mistakes and if you have any code you actually care about I would watch them like a hawk, in a nice large IDE on the side. The mistakes have changed a lot - they are not simple syntax errors anymore, they are subtle conceptual errors that a slightly sloppy, hasty junior dev might do. The most common category is that the models make wrong assumptions on your behalf and just run along with them without checking. They also don't manage their confusion, they don't seek clarifications, they don't surface inconsistencies, they don't present tradeoffs, they don't push back when they should, and they are still a little too sycophantic. Things get better in plan mode, but there is some need for a lightweight inline plan mode. They also really like to overcomplicate code and APIs, they bloat abstractions, they don't clean up dead code after themselves, etc. They will implement an inefficient, bloated, brittle construction over 1000 lines of code and it's up to you to be like "umm couldn't you just do this instead?" and they will be like "of course!" and immediately cut it down to 100 lines. They still sometimes change/remove comments and code they don't like or don't sufficiently understand as side effects, even if it is orthogonal to the task at hand. All of this happens despite a few simple attempts to fix it via instructions in CLAUDE . md. Despite all these issues, it is still a net huge improvement and it's very difficult to imagine going back to manual coding. TLDR everyone has their developing flow, my current is a small few CC sessions on the left in ghostty windows/tabs and an IDE on the right for viewing the code + manual edits.

Tenacity. It's so interesting to watch an agent relentlessly work at something. They never get tired, they never get demoralized, they just keep going and trying things where a person would have given up long ago to fight another day. It's a "feel the AGI" moment to watch it struggle with something for a long time just to come out victorious 30 minutes later. You realize that stamina is a core bottleneck to work and that with LLMs in hand it has been dramatically increased.

Speedups. It's not clear how to measure the "speedup" of LLM assistance. Certainly I feel net way faster at what I was going to do, but the main effect is that I do a lot more than I was going to do because 1) I can code up all kinds of things that just wouldn't have been worth coding before and 2) I can approach code that I couldn't work on before because of knowledge/skill issue. So certainly it's speedup, but it's possibly a lot more an expansion.

Leverage. LLMs are exceptionally good at looping until they meet specific goals and this is where most of the "feel the AGI" magic is to be found. Don't tell it what to do, give it success criteria and watch it go. Get it to write tests first and then pass them. Put it in the loop with a browser MCP. Write the naive algorithm that is very likely correct first, then ask it to optimize it while preserving correctness. Change your approach from imperative to declarative to get the agents looping longer and gain leverage.

Fun. I didn't anticipate that with agents programming feels *more* fun because a lot of the fill in the blanks drudgery is removed and what remains is the creative part. I also feel less blocked/stuck (which is not fun) and I experience a lot more courage because there's almost always a way to work hand in hand with it to make some positive progress. I have seen the opposite sentiment from other people too; LLM coding will split up engineers based on those who primarily liked coding and those who primarily liked building.

Atrophy. I've already noticed that I am slowly starting to atrophy my ability to write code manually. Generation (writing code) and discrimination (reading code) are different capabilities in the brain. Largely due to all the little mostly syntactic details involved in programming, you can review code just fine even if you struggle to write it.

Slopacolypse. I am bracing for 2026 as the year of the slopacolypse across all of github, substack, arxiv, X/instagram, and generally all digital media. We're also going to see a lot more AI hype productivity theater (is that even possible?), on the side of actual, real improvements.

Questions. A few of the questions on my mind:
- What happens to the "10X engineer" - the ratio of productivity between the mean and the max engineer? It's quite possible that this grows *a lot*.
- Armed with LLMs, do generalists increasingly outperform specialists? LLMs are a lot better at fill in the blanks (the micro) than grand strategy (the macro).
- What does LLM coding feel like in the future? Is it like playing StarCraft? Playing Factorio? Playing music?
- How much of society is bottlenecked by digital knowledge work?

TLDR Where does this leave us? LLM agent capabilities (Claude & Codex especially) have crossed some kind of threshold of coherence around December 2025 and caused a phase shift in software engineering and closely related. The intelligence part suddenly feels quite a bit ahead of all the rest of it - integrations (tools, knowledge), the necessity for new organizational workflows, processes, diffusion more generally. 2026 is going to be a high energy year as the industry metabolizes the new capability.

[MFH @fardeeze]

@jimcramer Altseason

[Jurrien Timmer @TimmerFidelity]

The question on everyone’s mind is whether this market has gotten over its skis betting that the AI boom will lift the economy’s non-inflationary speed limit and, in the process, enabling the US to outgrow an increasingly problematic debt burden.

I did a back-of-the-envelope exercise using the discounted cash flow model (DCF) to estimate at what combination of earnings growth and equity risk premium (ERP) the current level of the S&P 500 may be justified. Looking at the Bloomberg earnings estimates for the next few years and assuming a trend-like growth rate of 7% thereafter, the projected 5-year compound growth rate (CAGR) for earnings is 11%.

Given that 11% projection, the current level for the S&P 500 index only makes sense at an implied equity risk premium (iERP) of 3.7%. That’s what the DCF model solves for: it assumes the price is correct, and then using variables like earnings growth, the payout ratio, and the risk-free rate, it spits out a “required return.” From that return we subtract the 10-year yield and we get the iERP. The current iERP of 3.7% is not the lowest ever but as the chart shows below it's well below the long-term average of 5%.

We can use the DCF to solve for earnings growth as well, by holding the ERP constant. If the ERP was currently at its historical average of 5.0% instead of 3.7%, and we take that 11% EPS expected growth rate and plug it into the DCF model, the S&P 500 fair value drops to roughly 5,000. Ouch.

Putting this another way, at the current price level of 6730 and a “normal” ERP of 5.0%, we would need to see earnings grow by 18% per year for the next 5 years to justify the market’s current price level. It’s not impossible if the AI boom is the game changer that many expect it will be, but it’s not a small feat either.

Anyway, this kind of DCF exercise is a good way to assess what the market needs to believe for current levels to make sense. Currently it seems to believe that earnings will substantially surprise to the upside. The market is priced for success.

[Ray Dalio @RayDalio]

https://t.co/BytVjrzXkL

[MFH @fardeeze]

Defining 72hrs-1wk: Oct29 Fed cuts (inflation/U/E met, 1st BS expansion since ’21). Oct31 $31B BTC opts expire (~270k BTC), MMs hedge w/ spot - Exch reserves 2M lows → squeeze. Gold topped, $19T available for rotation to equities/crypto. Stables +0.5%/day. Floor $111k, max pain $120k. BTC enroute to $150k. Nov silent? Reassess all.