Ognyan Genev

recommended reading build.ms/2026/4/1/the-c…

@badlogicgames Do you use high or xhigh?

i have real trust issues with opus. it's extremely weird. any time it suggest a thing, i switch to gpt and ask for a second opinion. in +60% of cases, gpt finds additional things opus just missed. (not a gpt shill, love opus, but it's complicated)

@doodlestein man, please prioritize quality over quantity…

I’ve been a bit bummed recently about how my output has fallen off from a few weeks ago, when I was routinely pushing over 1,000 commits a day. But even 500+ is still a lot (plus I think it’s mostly because I’m making fewer, larger commits now).

@steipete @JustinGorya why not Pi?

@JustinGorya the app's amazing for knowledge work, emails, slack, cal, notion, linear, github, smaller fixes. For deep coding work it's hard to break old habits.

BREAKING! Peter is still using CodexCLI instead of the Codex App. He is still on the good side!

@badlogicgames Good read, when I see people working on 5 projects concurrently and push hundreds of commits per day it is pretty obvious that they are not doing quality work and I don’t want to use what they are building. Yes, it is skill issue, I know, but it is also skill to SLOW DOWN :)

I'm usually not one to write thought pieces without much technical depth. But here we go. Slow the fuck down. mariozechner.at/posts/2026-03-…

A bespoke software revolution? I don't buy it. It'll exist. It already exists. Small consultants and big consulting firms have made custom software for years. It almost always sucks. It’s bloated, confusing, and because the client pays, it’s built wrong in all the ways. Who’s excited about bespoke software? Software makers! Of course they're excited about building bespoke software — that's what they do. X is full of them. Your feed is full of people who love making software talking about making software. Of course they’re excited about the revolution. Echo, echo, echo... Most people don’t like computers. Nobody in tech wants to say that out loud. People tolerate computers. They use them because they have to. Given the choice, most would rather not think about them at all. So when someone suggests that AI means everyone will build their own custom tools, ask who "everyone" is. The three-person accounting firm drowning in client paperwork? They want the paperwork gone, not a new system to maintain. The regional logistics company with 40 trucks? They want the routes optimized, not Joe spouting off about this new system he’s been messing around with. The law firm billing 70-hour weeks? They want leverage on their time, not a software project to design. They don’t hate technology. But building and maintaining their own critical systems isn’t their wheelhouse, regardless of how much faster and easier it’s become. It's another job on top of the job. Will these people use AI? Absolutely, for all sorts of things. Will some outliers go deep and build real custom systems? Sure, but they're almost always people who already had some pull toward software. The curiosity was already there. They were dabblers before. Giving everyone access to software building tools doesn't mean everyone becomes a builder. A powerful excavator doesn't turn a homeowner into a contractor. Most people just want the hole dug by someone else. They don’t want the responsibility either.

@pvergadia @openclaw codebase is the real world example here, let’s see how easy it is to fix a bug without introducing a regression 🤣

🤯BREAKING: Alibaba just proved that AI Coding isn't taking your job, it's just writing the legacy code that will keep you employed fixing it for the next decade. 🤣 Passing a coding test once is easy. Maintaining that code for 8 months without it exploding? Apparently, it’s nearly impossible for AI. Alibaba tested 18 AI agents on 100 real codebases over 233-day cycles. They didn't just look for "quick fixes"—they looked for long-term survival. The results were a bloodbath: 75% of models broke previously working code during maintenance. Only Claude Opus 4.5/4.6 maintained a >50% zero-regression rate. Every other model accumulated technical debt that compounded until the codebase collapsed. We’ve been using "snapshot" benchmarks like HumanEval that only ask "Does it work right now?" The new SWE-CI benchmark asks: "Does it still work after 8 months of evolution?" Most AI agents are "Quick-Fix Artists." They write brittle code that passes tests today but becomes a maintenance nightmare tomorrow. They aren't building software; they're building a house of cards. The narrative just got honest: Most models can write code. Almost none can maintain it.

@bradmillscan When you realise that OpenClaw is LLM that puts a bunch of markdown files into context and try to predict the next token based on probability…

350-400 hours into OpenClaw over the last 33 days non-stop, no days off...I'm ready to quit. My openclaw is fucking lost in the weeds every day today and it's driving me nuts. Basic shit. I asked it to use GitHub. it has a GitHub skill. We have a GitHub SOP. I can see it's thinking process about using skills, then narrating how the skill doesn't exist, then going and inventing ways to retrieve the capability to use GitHub from the internet. I tell it to look in the openclaw docs for the proper skill path, it says "oops my bad, yeah it was there after all." This is ChatGPT 5.4 with extra high thinking turned on. I ask it to diagnose the problem only, so it goes and sees the system prompt is telling it to look at the wrong place, and it goes to GitHub and opens a GitHub issue about this 'bug' without even asking me. What the actual fuck. 3 hours on a Sunday of trying to rewire the brain of my openclaw to do default-behaviour. This thing such a productivity suck & mental poison. I can't do anything useful or positive with OpenClaw because I'm nonstop fighting fires in the engine room. I'm thinking about giving up.

@kelvinfichter @AzFlin What framework do you use or do you build your own? I’m thinking about using ZeroClaw…

@AzFlin I joke, I get a large % of value out of managing just a few agents at a time, but I am investing that time into a compounding agent setup of my own and it is increasingly doing more of the work. Recommend building your own, lots of fun.

skill issue unfortunately

How many iterations of fixing/implementing with codex and then /review do you do before the review comes clean at least a few times in a row? My observation is that on a complex codebase (@openclaw, for example), it takes forever to fix an issue with codex 5.3 xhigh without introducing new issues... @steipete any hints?

its honestly the perfect time to be unemployed, way more hours tinkering with AI and @openclaw 🤖.

@dcbuilder not me

does anyone actually use Claude Cowork or Perplexity Computer as a dev?

@steipete what’s your take on nanoclaw, zeroclaw and other compact claws?

Today I played for the first time with @openclaw installed on a local VM on my MacBook Pro. After the initial setup, I asked it to create a new file on the desktop with a surprise for me. It is fun and a sneak peek into the future.

@mitchellh @steipete Before 5.2, Opus was the better model for me, but 5.2 xhigh was a gane changer and I never looked back to Claude since then.

I know this is pretty well established at this point, but Codex 5.3 is a much more effective model than Opus 4.6. I went back and forth on both for a bit, but haven’t touched Opus at all now for a full week. First model to get me off of Opus… ever. Good job Codex team.

I ignored high LDL/total cholesterol for years. I told myself the usual stuff: “I’m fit, I train, I eat clean, I’m probably fine.” Last week, my cardiologist ordered a CAC scan. My score came back 42 at age 40! So basically, I had the arteries of a 55-60-year-old man. That was a wake-up call, and I'm on statins now. If your cholesterol has been high for some time, ask your doctor if a CAC scan is suitable for you.

@brian_armstrong @base Reputation for AI agents: github.com/ogenev/trust-n…

What are folks building on @base this week? Will check out a few projects

@grok what do you think? Be very honest and critical.

I spent some time tinkering with AI agents, reputation, and Ethereum. An open-source, verifiable trust layer for AI agents on ERC-8004 seems very appealing. This is why I started to build TrustNet: github.com/ogenev/trust-n…. Any feedback is welcome. The contracts are deployed now on Base Sepolia, still playing with it, and looking for interesting ideas.