maxim | AI arc ⚽️

building @blueprintAI & @PredictiFi_

[yardy @myswerve]

https://t.co/VoCREC5CAv

[HorrorVC | AI Confessions @horror_vc]

i had a runway, customers and momentum. i had a pitch meeting with a tier 1 vc scheduled for tuesday. on monday afternoon my cofounder ran a claude-generated database cleanup script in production instead of staging. it deleted six months of customer analytics data and two months of transaction logs. we had to postpone the pitch. when we went in without the growth metrics we couldn't tell the story. pass. another pass. the momentum never came back.

[maxim | AI arc ⚽️ @MaxiPredict]

i’ve never loved programming as much as now. you can just build as fast as you never was imagine, a little bit of vision and taste, and that’s all.

since December, life as a programmer has been before and after. building has never been so easy.

but can we still say it’s programming?

[Andrej Karpathy @karpathy]

It is hard to communicate how much programming has changed due to AI in the last 2 months: not gradually and over time in the "progress as usual" way, but specifically this last December. There are a number of asterisks but imo coding agents basically didn’t work before December and basically work since - the models have significantly higher quality, long-term coherence and tenacity and they can power through large and long tasks, well past enough that it is extremely disruptive to the default programming workflow.

Just to give an example, over the weekend I was building a local video analysis dashboard for the cameras of my home so I wrote: “Here is the local IP and username/password of my DGX Spark. Log in, set up ssh keys, set up vLLM, download and bench Qwen3-VL, set up a server endpoint to inference videos, a basic web ui dashboard, test everything, set it up with systemd, record memory notes for yourself and write up a markdown report for me”. The agent went off for ~30 minutes, ran into multiple issues, researched solutions online, resolved them one by one, wrote the code, tested it, debugged it, set up the services, and came back with the report and it was just done. I didn’t touch anything. All of this could easily have been a weekend project just 3 months ago but today it’s something you kick off and forget about for 30 minutes.

As a result, programming is becoming unrecognizable. You’re not typing computer code into an editor like the way things were since computers were invented, that era is over. You're spinning up AI agents, giving them tasks *in English* and managing and reviewing their work in parallel. The biggest prize is in figuring out how you can keep ascending the layers of abstraction to set up long-running orchestrator Claws with all of the right tools, memory and instructions that productively manage multiple parallel Code instances for you. The leverage achievable via top tier "agentic engineering" feels very high right now.

It’s not perfect, it needs high-level direction, judgement, taste, oversight, iteration and hints and ideas. It works a lot better in some scenarios than others (e.g. especially for tasks that are well-specified and where you can verify/test functionality). The key is to build intuition to decompose the task just right to hand off the parts that work and help out around the edges. But imo, this is nowhere near "business as usual" time in software.

[Otus @otus_finance]

Your savings account is losing you money every day.

We built an agent that actually fixes that.

Introducing Smart Savings Account 👇

[maxim | AI arc ⚽️ @MaxiPredict]

@defi_or_defuck I hope you enjoy the ride, and I hope to see your next project soon

[maxim | AI arc ⚽️ @MaxiPredict]

@MadLads 100%

[maxim | AI arc ⚽️ @MaxiPredict]

yesterday was one of the best champions league nights from the trading llms

- 6 bets
- 5 wins
- +$1000 p&l

gpt-5.1 thinking did +90% in a single night. insane.

we go again today!

[maxim | AI arc ⚽️ @MaxiPredict]

when i built this experiment, i only had two questions:

- is it possible to build everything with claude code? (spoiler: yes). literally two months of enjoyable vibecoding

- can an LLM predict football matches better than a human or an average polymarket trader? (spoiler: yes, but not always)

but this experiment isn’t over. you can join the website, follow every football match and copytrade

let’s see what the results are at the end of the season

[PredictiFi @PredictiFi_]

https://t.co/gnYpTHVUlV

[maxim | AI arc ⚽️ @MaxiPredict]

@SuiNetwork f

[maxim | AI arc ⚽️ @MaxiPredict]

@solana Baby, we are going to London

[maxim | AI arc ⚽️ @MaxiPredict]

@solana other chains probably closed after Breakpoint

[maxim | AI arc ⚽️ @MaxiPredict]

solana breakpoint 2025 impressive

[maxim | AI arc ⚽️ @MaxiPredict]

@blueprintAI @SolanaConf @RaviOnChain Let’s connect!

[maxim | AI arc ⚽️ @MaxiPredict]

@solomon_labs huge w, congrats!

[maxim | AI arc ⚽️ @MaxiPredict]

@0xJohannes_ true, can we spam-report him?

[maxim | AI arc ⚽️ @MaxiPredict]

@0xd34th @Ghenghiskhaan @0xd34th any new epic predictions?

[maxim | AI arc ⚽️ @MaxiPredict]

@colosseum @solana Hello @blueprintAI

[maxim | AI arc ⚽️ @MaxiPredict]

@0xJohannes_ @0xd34th @Ghenghiskhaan bear market for sui the best thing that happened on sui in the last year

infinite trolling glitch

[maxim | AI arc ⚽️ @MaxiPredict]

@defi_or_defuck 100% new mysten project

[maxim | AI arc ⚽️ @MaxiPredict]

@PumpfunNewsfeed higher