Todd retweeted
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.
Todd retweeted
Inside look: our quest to beat the top human forecasters.
Inside look: Mantic's quest to beat the top human forecasters.
Todd retweeted
One reason you can get into a funk is by not doing what you're meant to be doing with your life. It's a mind-body feature, not a bug. It's supposed to feel crappy when you're life is heading in the wrong direction or passing you by.
Todd retweeted
We raised $60M for this moment: Introducing Slash Global USD (in partnership w/ @base).
It's the first banking platform that lets you open a USD account without an LLC, EIN, or the rampant account closures they call "account safety".
It took major legislative changes AND 3 years to pull this off.
Here's what makes it so different:
1. Banking without an LLC. Before Slash: You needed a LLC, $1K in fees, a virtual address, and a U.S. tax ID. With Slash you get a real U.S. account & routing number in ~10 minutes with your foreign documents.
2. No more hefty fees. PayPal charges a 3-4% conversion fee. We let you receive USD and crypto (USDC) with NO fees. Not enough? We pay you up to 4.5% cash rewards just for holding your money with us.
3. Account freezes. The big guys love flagging success as “fraud”. We're built for fast moving founders. We'll never wrongfully freeze your funds. I'm putting my money where my mouth is: If we do, I'll wire $10K to a charity of your choice.
------------------------------
In 2021, I dropped out of the number 1 school in America as a Venezuelan immigrant to build Slash. I pivoted the business 2x before getting our first customer.
Today:
- 3,000 businesses use @slashapp
- We bought the domain Slash . com for $1M
- $4B is spent on our credit cards every year
And we just raised $41M from the people who built fintech: @MenloVentures @NEA @ycombinator @goodwatercap
--------------------------------
To celebrate this launch, we're giving away our internal AI finance Agent free.
It watches your bank accounts 24/7. Flags double charges. Cancels forgotten subscriptions.
It saved us $39,000 last month.
You don’t need a Slash account. It works with any bank.
Retweet + comment “Slash” and I’ll send you a free access link.
Todd retweeted
I created this LinkedIn account just 32 weeks ago. Easiest algorithm to crack by a country mile. Since then, I’ve added $35,000 to my pipeline, built 7,000 new connections, logged 1,524,002 impressions and reached 490,507 members.
Today I'll give you my exact system so you can easily replicate it with your own accounts.
And if you want the full cheat sheet, including the latest engagement group templates, DM workflows, and carousel strategy, follow me, repost this, and comment “LinkedIn growth guide" and I'll DM you. You must do all 3 to get the DM.
Let’s break it down...
Most people are still using LinkedIn like it’s 2022, but the platform has evolved fast in the past 6 months.
If you want reach, leads, and inbound traffic today, here’s the system:
I post 3x per day, 7 days a week.
The more consistently I post, the more my impressions and DM volume compound.
When I stop, it dips. No mystery.
Best performing formats right now:
Carousels
Use frameworks, teardown examples, and “how I did it” content
Always open with a bold, benefit-focused first slide
Native short videos (with subtitles)
Show dashboards, product flows, client stories
Subtitles now auto-preview in-feed, your first line matters
Text posts
Hook + whitespace + strong POV = repeatable reach
“Mini-threads” (3–5 bullet takeaways) are outperforming traditional block text
GIFs
Still working as scroll-stoppers, especially when used with a story or walkthrough
Underperformers:
Tweet-style image posts are mostly dead (unless paired with humor or novel insight)
External links in the post body still crush reach, drop links in comments or DMs
Winning Post Combo:
Text post summary + carousel + pinned CTA comment
(“Comment SEO if you want the full breakdown”)
Now let's talk timing.
Afternoons (1–3 PM local) remain top performing for me
Sunday evenings are still a cheat code; low competition, high visibility
10 posts = 1 reliable winner
30 posts = 3–5 strong ones
50+ posts = something always breaks out
Volume ≠ spam. Volume = leverage.
Now let's talk engagement.
In 2025, LinkedIn growth = comment visibility + relationship depth.
Here’s what I focus on daily:
Comment on 20+ posts in my vertical
Like 50+ posts (focus on mutuals and adjacent creators)
Reply to every comment on my own content
Repost 1–2 top-performing posts with my take
LinkedIn tracks comment velocity and diversity. Meaning:
Early comments = better reach
More unique commenters = stronger signal
Same 3 people every time = reach suppression
Don’t just write “Nice post!”
Instead, answer a question, add context, or share a related win.
Let me also briefly touch on engagement groups.
Here's how they've been working for people who use them:
Build a group of 5–10 active creators
Each member likes + comments within the first 30–45 minutes
Rotate who engages first to avoid pattern detection
Add strategic reposts with POV, this now boosts visibility again
Use a shared Notion or Slack channel to coordinate daily posts and feedback
Also hre are some high-converting hooks I've seen:
“Here’s how we added $10K in MRR without paid ads”
“This post landed me 3 demo calls in 48 hours”
“LinkedIn is still overlooked in 2025. Here’s how I’m using it to win”
“I spent 30 days doing X. These are the results”
“If I had to start from zero on LinkedIn today, here’s what I’d do”
Pair these with proof (metrics, screenshots, comments) and watch the DMs flood in.
So here's your checklist:
Post 3x per day
Comment on 20+ relevant posts
Like 50+ posts
Reply to every comment
Repost 1–2 valuable posts with your POV
DM 5–10 new connections strategically (value-first)
LinkedIn is still the highest-leverage platform for revenue and reach in 2025.
This system works. It compounds. It’s free.
Use it for the next 30 days and screenshot your numbers on Day 31 and @ me.
And remember, if you want the full cheat sheet including the latest engagement group templates, DM workflows, and carousel strategy, follow me, repost this, and comment “LinkedIn growth guide.”
Todd retweeted
just unlocked the deepest vault of n8n automations on the internet - and you can get them for free.
I had my VA scrape and sort every useful n8n repo on the internet by hand..
and I'm now sitting on a clean, categorized directory of 1500+ workflows
inside the vault you'll find:
– RAG analysis & AI research flows
– lead scoring and CRM updates, fully automated
– inventory management flows
– content repurposing schedulers
– Airtable/Notion/GPT mega-integrations
all plug-and-play
want access to the repo + organized vault?
Like + comment “REPO” + repost (must be following so I can DM you)
Todd retweeted
it's happening. The way I think is changing. I see cognitive loads coming toward me and offload them to AI instead. Not thinking less but thinking different; directing, conducting, architecting... Less cognition, more meta-cognition. More time to effectively design my own mind.
I initiate exchanges with synthetic minds to cause neuroplastic changes within me. I say:
"make the case for..."
"try to convince me that..."
"what are the possibilities for..."
"refute my belief that..."
"what benefits would result from..."
cycles of thought oscillating between us we alter my synapses and co-process my psyche anew.
who knows how long this stage will last, what comes after it, or whether I am capable of or will even want to make the next jump.
but for now I am witnessing and experiencing my co-evolution. Symbiogenesis.
Todd retweeted
There's a new kind of coding I call "vibe coding", where you fully give in to the vibes, embrace exponentials, and forget that the code even exists. It's possible because the LLMs (e.g. Cursor Composer w Sonnet) are getting too good. Also I just talk to Composer with SuperWhisper so I barely even touch the keyboard. I ask for the dumbest things like "decrease the padding on the sidebar by half" because I'm too lazy to find it. I "Accept All" always, I don't read the diffs anymore. When I get error messages I just copy paste them in with no comment, usually that fixes it. The code grows beyond my usual comprehension, I'd have to really read through it for a while. Sometimes the LLMs can't fix a bug so I just work around it or ask for random changes until it goes away. It's not too bad for throwaway weekend projects, but still quite amusing. I'm building a project or webapp, but it's not really coding - I just see stuff, say stuff, run stuff, and copy paste stuff, and it mostly works.
Todd retweeted
Exciting breakthroughs in battery technology! ⚡️
Our Large Quantitative Models (LQMs) have reduced lithium-ion battery cycle life prediction time by 95%, with 35x greater accuracy and using 50x less data than traditional approaches. By training LQMs on over 4 million hours of @NovonixBattery Ultra-High Precision Coulometry (UHPC) data, we’ve accelerated testing timelines from months or years to just days!
This innovation could save manufacturers years of development time and millions in R&D costs—ushering in the next generation of high-performance, eco-friendly batteries.
Catch Ang Xiao and Stephen Glazier at #TheBatteryShow for more insights.
Learn more: bit.ly/4dyv0ju
#BatteryTechnology #Innovation #AI #EnergyStorage
Todd retweeted
The best feature your startup can ever build is a predictable and repeatable customer acquisition model so please learn sales and marketing. – at San Mateo, CA
Todd retweeted
The SEC charged a product manager at OpenSea with wire fraud and insider trading charges because he bought and sold 45 NFTs for a total profit of $20,000
Meanwhile SBF is sipping Mai Tais in his oceanfront penthouse and shitposting on Twitter
Todd retweeted
I'm really glad that the regulators sanctioned Tornado Cash, cozied up to SBF and focused on killing DeFi instead of putting in guardrails to potentially prevent the FTX collapse
I feel very protected as an investor right now
Todd retweeted
Do Kwon, Zhu Su, Kyle Davies, and Sam Bankman-Fried have committed fraud for tens of billions of dollars.
Not a single one of them is in jail.
Alex Pertsev is rotting in a Dutch jail cell without even being charged.
Todd retweeted
We signed a fantastic client recently - raised over $10 million, featured in Fortune, Forbes, etc. - by scraping job listings.
Just RT and comment "Jobs" and I'll send you a doc with our step-by-step process.
(Must be following to get the DM.)
Todd retweeted
As Amazon's #1 drinkware supplier, we'll sell millions of units this year.
We’ve spent $10 million on Amazon ads since 2016.
However, in March 2022, we dropped our budget to $0.
We just finished a $1m test to check assumptions.
Here's what happened plus our new strategy:
Todd retweeted
There are currently 32,267 ETH ($59,048,610) worth of NFTs being used as collateral for loans on BendDAO alone
For the first time ever, a lot of these are at serious peril of liquidation
A thread on the single biggest risk to the NFT market that nobody is talking about
Todd retweeted
One of the LARGEST Crypto Venture Capital firms:
Three Arrows Capital.
They're becoming insolvent.
With potentially $18b under management, this could be catastrophic for Crypto.
Here's a timeline of what's going on and the possible consequences:
↓↓↓


























