Duco (@Ducotr) header
Duco (@Ducotr) profile picture

quantimatician and pro chef

Joined July 2018
Photo by Duco (@Ducotr)Photo by Duco (@Ducotr)Photo by Duco (@Ducotr)Photo by Duco (@Ducotr)Photo by Duco (@Ducotr)Photo by Duco (@Ducotr)Photo by Duco (@Ducotr)Photo by Duco (@Ducotr)Photo by Duco (@Ducotr)Photo by Duco (@Ducotr)
you dont work on bleeding edge stuff if your slack isnt littered with this guy
78
108
37
2,083
200,468
nån som ska till slush helsingfors?….. kom dm
3
372
Claude nere it’s over
2
720
Typ så jag kör också funkar bra för det mesta
A lot of folks ask me "how do you work with agents in multiple repos at once?" I have a very simple solution - the Home Repo I have a single repo (the home) where all of my skills, steering files, wiki etc live. I run my agent in the home repo, and then inside AGENTS.md I have pointers to the other repos: ``` For skills, clone mattpocock/skills to ~/repos/skills. For Sandcastle, clone mattpocock/sandcastle to ~/repos/sandcastle ``` This is extremely simple and incredibly effective. The home repo lets you: 1. Put all your steering files and tooling in one place 2. Have a single spot to run all your agents in 3. Make sweeping changes over multiple repos Talked about this in the course yesterday and folks are loving it (see below). I'm sure you will too. Enjoy!
1
318
new me
1
27
680
✍️
The benefits of abundant intelligence will accrue far outside of San Fransisco. Spent the last year writing about capital, sovereignty, and the coming reordering of economic power. THE COMMANDING HEIGHTS Commanding-Heights.com
1
329
Bra takes Fokusera på att hitta saker som kan testas i kod så agenten kan hillclimba Lägg tiden på att skriva bra specs Om du inte vet var du ska börja så testa detta från Google: opensource.googleblog.com/20… Be Claude implementera men kanske skippa swarms till o börja med
I interviewed ~2 dozen researchers, engineers, research / engineer managers etc. asking them how they / their teams use coding agents in their work. Then I sprinkled it with my opinions and hot takes and wrote a substack post (link in the next tweet, I'd be delighted if you subscribe; of course, the post is free and any future post will be free too!) trying to triangulate the current reality of software engineering / industry research as opposed to what gets peddled on social media. TL;DR (3. and 4. are the most important!): 1. Models, harnesses, workflows: don't bother with fancy workflows, they don't really work. Use stock Claude Code / Codex (or Copilot with Claude / GPT models if you are a Microsoft stan), perhaps with subagents in some very special cases. Write a design doc, then plan a change, then implement it, then merge. Wash, rinse, repeat. Don't get swayed by people online who are trying to sell you snake oil in the form of their new harness or workflow. None of this stuff works (in general). 2. Correctness/tests/reading or not reading code: it's OK not to read code if you do one of the following three things. a) One-off stuff / demos, which you need to work on exactly one input. b) Testable / verifiable stuff, e.g., GPU kernels. c) Stuff, where it's OK to be "directionally correct" (ML / data work / research code). In other cases, especially if you require any degree of maintainability / reliability or your code is customer-facing, reading code is still necessary. Reading tests and steering the model to write better tests is a good idea too. It's often good to use a different model to review the code that the other model wrote. Even if you approve the PRs without changing them, reading still helps for future more productive prompting. 3. Team dynamics: if we are not careful, agents can really undermine team dynamics. Two biggest problems are "slop infestation", where over time all PRs / emails / Slack messages end up being unreadable slop, and "overbuilding", where we are building and shipping useless stuff to get recognition or as a form of procrastination. Other, less severe but still important problems include: reduced communication, being secretive about ideas, decreased diversity of thinking, mode collapse of the ideas, trusting LLMs more than people. 4. Call for action: we really need to be careful with preserving team work and collaboration. We still need to "form packs to hunt mammoths", but with coding agents this becomes increasingly harder. Overall, many of the agents-induced problems stem in myopic short-term thinking of the sort "we have a couple of years left until our jobs are fully automated, so I can as well extract as much value as possible, everyone else be damned". While somewhat understandable, I think it’s important to resist these urges, if nothing else, for your own sanity’s sake. [Full disclosure: until 2 months ago, I worked at OpenAI and my work directly contributed to Codex, I'm a big fan of coding agents and quite strongly AI-pilled, but there is little I hate more than fearmongering, gaslighting, lies and rage-bait that are ubiquitous in this space, unfortunately, hence spending time on all of this.]
1
7
903
New blog post - Chasing the Karpathy Peak (and Falling Off Slop Cliff): Confessions from Slopaholics Anonymous My story of becoming a slopaholic, who’s most at risk, and how to avoid it (or recover) thomasjiralerspong.substack.…
40
128
52
941
147,698
Köp bara Claude max 100 eller 200 dollar. Finns inget mer prisvärt nu.
Replying to @fsporsen
vilken är den mest prisvänliga AIn?
2
20
6,222
Säg att haiku 5.5 subagents ska köra 100k token context som max då är den 5x billigare än annars
1
2
588
med denna setup kommer det vara svårt att ta slut på 5h fönster på 100 dollars på ett projekt med minimal förlust jmf med bara köra opus 5.5
1
434
onda är de anthropic
Asked Claude to read the Anthropic S-1 and summarize it for me Claude proceeds to tell me it accidentally sent my email to the SEC
411
mkt sant även för vanlig utveckling
u may not like it but the apex prompts are 3-5 lines full of typos, swearing/caps, random proprietary technology names and lies to make the model think the thing you are copying or breaking is yours
1
732
bajs
Wake up. Every profession deemed white-collar, the middle class dream, is about to be decimated by AI, at a much faster pace than anticipated, with monstrous social consequences.
7
1,834
efter opus 5.5 känns det lite såhär den hanterar det mesta hyfsat rätt svårt att inte känna ett behov att sträva mot ett anpassat ”duco os” dock men lätt att man fastnar i underhåll/workflow förbättringar ist för att driva resultat
🤖 Made with AI
Replying to @dumbodax
Hermes är typ det som openclaw ville vara Inte sett nån jag tycker verkar vettig som föredrar openclaw Men tycker fortf man hamnar i en loop av att ständigt behöva underhålla/förbättra/göra ändringar i setupen så gott ifrån hermes som daily driver
1
10
750
man e ju dock inget fan av anthropic egentligen riktiga råttor är de o alla jävla permissionsstrul hela tiden låt mig kalla min clanker för clanker
1
3
211
fick inge invites till top secret grupp… rip
finns de nå gruppchattar discords etc… där folk delar erfarenheter av ai implementationer etc testar nya verktyg osv Inge boomer/normie bullshit vad jag kör/kört nyligen: hermes omp/tern claude code tyvärr pga för generösa usage limits o opus 5.5 för bra uniichat/optchat
3
8
1,544
finns de nå gruppchattar discords etc… där folk delar erfarenheter av ai implementationer etc testar nya verktyg osv Inge boomer/normie bullshit vad jag kör/kört nyligen: hermes omp/tern claude code tyvärr pga för generösa usage limits o opus 5.5 för bra uniichat/optchat
9
1
28
7,323
big
Bend 2 is here! It is a new programming language that blocks AI mistakes via *proof checking* - the same technique big AI labs used to solve open math problems, like Navier-Stokes. It is also very fast, and runs on GPUs. Watch the video. Link in the comments.
2
731