Practical product and Al thinking for busy builders.

Chester, England
Joined February 2011
Anthropic employees expect AI to handle 75% of their current work hours by 2030. The average US worker expects AI to handle just 28%. That is a remarkable gap between people working at the frontier of AI and the broader workforce. These are expectations, not established outcomes. However, they raise an important question for anyone building a career today: If the people closest to AI expect this much automation, how seriously should the rest of us prepare?
1
2
2
578
niko retweeted
i feel like swe interviews could probably collapse into just 2 technical rounds: • system design (how well can you articulate your ideas, do you actually know how to engineer something) • onsite project to build a real thing with agents (can you translate your intent into high quality outcomes) everything else we used to ask is no longer necessary
🗞️ if your team is too busy doing their 'normal job' to experiment with AI, you're preparing them to be replaced ghuntley.com/replaced/ In this post, I guess this's the last time I'll say it. AI use is no longer optional if you wish to be employed. What follows is advice for current employees and hiring managers on how to hire in 2026 to source AI-first candidates.
292
414
113
6,455
423,365
RIP soft PRD templates sold as “AI-ready.” /make-requirements-great is free, sits in Claude, and beats your PRDs into shape Here’s what it does:
2
15
101
22,538
niko retweeted
cursor projects are just too good with pstack. i now routinely have at least 10 projects running in parallel, tackling everything from perf work, to tech debt clean up, experiments with Bend2 and rust, fixing user feedback, building new dashboards, building games i feel like a 1000x engineer cursor.com/blog/projects
177
68
21
1,813
123,856
Another insane Jev use case! Jev is making it dramatically cheaper to evaluate what actually happened inside an agent run. And finally, someone open-sourced a self-improving memory layer that can put that signal to work across agent harnesses: - Claude Code - Codex - Cursor - OpenCode, and 20+ more Beacon by @asymptotelabs continuously captures your agent history across harnesses and uses Jev to identify which runs are actually worth learning from. It then turns the highest-signal workflows, corrections, and debugging patterns into reusable skills. GitHub repo: github.com/Asymptote-Labs/ag… (don’t forget to star it ⭐ ) Beacon preserves the complete session history. But preserving a run and learning from it are two different things. Most coding-agent sessions contain routine exploration, failed commands, and fixes that only apply to one task. The trace can remain available for inspection without turning every detail into guidance for future agents. Jev scores each run for evidence, reuse potential, and human correction signals. An application policy then decides whether to promote, review, or discard it. The recording shows this in action. Claude receives a coding task, modifies the implementation, and runs the tests. I then provide an edge-case correction, so Claude updates the code and adds regression coverage. Beacon automatically captures the complete session. Jev evaluates whether the correction contains a reusable engineering lesson. Once approved, that lesson becomes available to other coding agents working on the project. Since it works across harnesses: - Claude Code sessions can teach Codex. - Cursor debugging can improve OpenCode. So a problem solved by one agent should not need to be learned from scratch by another. If you want to dive deeper into Jev, I also wrote a hands-on guide to building this Jev-style decision path with open models, entirely locally. Read it below.
86
238
18
2,027
293,274
niko retweeted
Jev (@typesafeai) is so insane & cheap for search!! > 6000+ @ycombinator Startups indexed. > Sub 1 second search results. > 90M tokens & $2.7 in total testing costs. Search any startup in a second, in any way! - Color - Niche - Your Competitor - Age - Image - etc... > watch the entire video, it's so freaking cool omg! > this is the coolest thing i have ever built for fun! (worked on it for 2 days straight!)
82
56
18
1,699
130,720
Macadam is one of the best examples of what a top tier app onboarding should look like. Their onboarding flow is exceptionally well designed. Key things: - Engaging animations and vibrant visuals - Gradually collects information from the user - Clearly explains how the app works - Asks for permissions at exactly the right moments - Delivers instant dopamine with coins, rewards and streaks It barely feels like onboarding. It feels like you’re already using the product from the very first tap
This habit tracker app built by indie dev @edwin1 is one of the most impressive apps I’ve seen on the App Store. I absolutely love the onboarding. Onboarding - Fully customizable character - Medieval RPG theme - Start your first quest - Earn your first XP - Fight your first battle Paywall - Yearly subscription highlighted - 3 day free trial on the annual plan - Monthly and lifetime options available Edwin took a “saturated niche” like habit tracking and built a fully gamified RPG experience around it. That’s the key: take something that’s already validated and build something awesome and different around it
12
15
2
361
37,691
Jevmaxxing for marketers Jev can speed up most marketing workflows 30x and do it for < $3: 1/ Scan the whole Meta Ad Library -> It reads every live ad in your category and tags each one by hook, format, offer and days running 2/ Find the ad patterns that survive -> It compares formats by how many ads are still live after 60 days, so you know what lasts before you test it 3/ Score briefs before you shoot -> Your LLM writes the briefs, Jev scores each on hook, brand fit and survival odds, and only the top ones get made 4/ Sort search terms -> It asks "is this query from a buyer?" across the full Google Ads report, so negatives land the same night 5/ Catch fatigue early -> For every ad with frequency up and CTR down, it picks replace, refresh or leave 6/ Check ad to landing page match -> It scores whether the page delivers what the ad promised, the cheapest CVR fix in most accounts 7/ Score every lead -> It rates each form fill 0 to 100 against your ideal customer within seconds, so Google and Meta learn to find more of the good ones Available in the Ryze AI app and MCP/Claude Connector, link in the 1st comment 👇
Jev dropped the price of SEO/GEO fixes by 90% Agents that audit and fix a client's SEO/GEO used to cost us ~$250 Here's where the savings come from: 1/ 30x faster reads of Search Console and PostHog/Mixpanel data 2/ 30x faster checks of what ChatGPT searches on Bing 3/ 30x faster modeling of what users ask Gemini and Claude 4/ 30x faster scans of who ChatGPT and Claude cite 5/ 30x faster analysis of the sources behind those citations 6/ 30x faster gap analysis: why they get cited and we don't 7/ 30x faster fixes across 1,000s of pages on large client sites 8/ 30x faster sorting of which page types ChatGPT cites 9/ 20x faster creation of the pages that make ChatGPT pick you Available in the Ryze AI app and MCP/Claude Connector, link in the 1st comment 👇
50
107
17
1,397
321,062
Easy million dollar business. Make it so I never have to re-enter my information at a doctor's office again. I'd pay a lot for that.
82
11
5
313
26,231
niko retweeted
Jev + GrokBot is the best AI agent system I’ve built in my life It just made my setup CHEAPER and FASTER than what 95% of people are running... setup takes literally 7 minutes: prompt → GrokBot → Jev decision → GrokBot execution → result step 1 → open @typesafeai , create API key (keep it off chat paste) step 2 → tell Grok Bot: store TYPESAFE_API_KEY in the secure field step 3 → prompt Grok Bot: install typesafe-sdk on Agent Computer + smoke system_one (Choice) step 4 → tell Grok Bot: build the usage lab (router, dry-run, config, logs) - or clone Github below step 5 → add skill jev-usage-router: before browser / research / retry / extra bot → call the router, honor action step 6 → stay shadow first, read logs, then active when you trust it - kill switch: bypass jev or enabled: false step 7 → flip active: GrokBot obeys route - Jev decides - GrokBot executes - humans control irreversible actions the result: Jev + GrokBot the best and fastest agent running directly on your computer rn, I’ve already tested it on routine tasks - and the results are genuinely incredible You can come up with endless ways to use Jev + GrokBot - but the most important thing is to install it as soon as possible Copy this 2028 setup, explore my repo below - then read the full Jev deep dive ↓
Jev is the "Internet" moment for the AI industry It tells your agents and LLMs what to do next, in milliseconds and at almost zero cost If you set it up correctly, you will have the AI engineer’s stack for 2028 In this article, I show you how
123
258
40
2,417
387,159
niko retweeted
Andrej Karpathy spent 8 years at OpenAI and Tesla Last week, he condensed everything he knows into one free 2-hour lecture Agents → Loops → Harness → Self-Improving Systems People pay $14K for bootcamps that teach less than this This lecture beats most paid AI engineering courses You probably don't have 2 hours right now Don't let this disappear from your feed Watch it Then read the article below
36
148
1
729
272,323
Jev Generative UI has been achieved externally
New experiment: json-render + jev The future Generative UI is instant Your components, your actions, your design system Rendered in milliseconds
98
155
20
4,468
533,887
I built a voice controlled computer-use for my mac using @typesafeai's Jev and it's INSANE how fast it is! I can dictate "open the notes app and create..." and the app opens before I even finish my sentence.
231
320
126
6,260
662,846
niko retweeted
Learn to fine tune a model this year Learn to fine tune a model this year Learn to fine tune a model this year Learn to fine tune a model this year Learn to fine tune a model this year Start small Start slow But make sure you start
Fine-tuning is about to become one of the most valuable AI engineering skills. Not because everyone needs a custom model. But because the people who understand how models learn from data will build things others can’t. Full guide:
70
527
19
6,663
925,585
Most predictions I see are still way too conservative. Here's mine
541
898
388
8,863
2,389,642
Jev just solved doomscrolling. You pick a niche, Jev reads 3 days of posts and asks 8 questions each. It runs in ~2s for $0.007 😅 > removes bait and hidden ads > judges the text and like/reply/repost ratios Free, no signup, go try it then touch grass ↓🌲
84
69
21
1,702
179,981
This is why I am on Twitter, to see posts like this
Solving a Rubik's Cube with graph theory.
195
18,376
98
226,259
6,566,215
This is the AWS moment for agents. OpenAI just shipped the Agents API, and everything that made agents hard to build is now something you rent instead of build. Keeping one running for days, remembering what it's doing, using tools, recovering when a step fails. Now it's an API call. Before AWS, you had to buy and run your own servers, so only well funded teams could build real/scalable products. AWS made the hard part rentable, and all the value moved up to whoever had the best idea for what to build on top. Same thing is happening here. I'll be honest, my first reaction was that this is bad news for a lot of people I know. If your whole company is an agent platform, the thing you spent the last year building is now included for the price of tokens. But the more I sat with it, the more I think it's really good news if you're building for one specific industry. Basically, vertical software type stuff. If the hardest engineering just became a line item, then the only thing left that's actually hard is knowing the job. Like what counts as a done claim, what makes a filing get rejected, where the money leaks. So the wedge now is owning one painful workflow with your own tools, data, approvals, and a clear ROI. Freight exceptions, insurance reviews, security triage, revenue leakage, healthcare admin, compliance ops. I see a lot of doomers on X saying software is over (and sometimes I feel the same way), you'll just use Claude or ChatGPT for everything. BUT, if that were true, OpenAI wouldn't be shipping infrastructure for other people to build agents on. They're telling you exactly what they think: the general model is theirs, the 10,000+ specific jobs it needs to be pointed at are yours. I think software is just evolving. It's becoming an agent internet, and every layer of it, the tools, the workflows, the boring jobs, gets rebuilt for a user that doesn't have eyes or hands. That's a lot of companies waiting to be built.
Go from idea to a working agent faster with the Agents API. Build and run cloud agents with the Codex harness, fully managed by OpenAI. We handle orchestration, long-running sessions, and context management. You focus on what makes your agent unique. Available in public beta.
176
150
28
2,149
429,759
this is f*cking wild that a whole hedge fund now fits in one 'pip install' a small team dropped a swarm of AI agents that runs the entire trading pipeline a fund's staff usually handles, and open sourced it you just wire it up and let it run: 1/ a director agent generates and validates the trading thesis, then hands it to a quant agent for technical and statistical analysis 2/ a risk agent sizes the position and checks it before anything touches the market 3/ an execution agent fires the order autonomously on Solana, with Coinbase support coming next every step logs a structured, auditable trail, built more like institutional infrastructure than a weekend script connect a wallet, and it trades on its own around the clock repo is open source and live on GitHub👇 follow for more
8
2
27
2,727
niko retweeted
as the operator of the loom is credited with the output cloth and the operator of the tank is credited with the kill as the LLM gets good, surely the one who brings its outputs to the people (in the manner of prometheus) must be credited with the resultant boons (or otherwise)
4
9
87
79,079