Andreas Keller

637 posts

Andreas Keller banner
Andreas Keller

Andreas Keller

@itsakeller

Web developer. Writing about Next.js

Thun, Switzerland Inscrit le Ekim 2013
1.8K Abonnements575 Abonnés
Andreas Keller
Andreas Keller@itsakeller·
@CasJam Wow that’s incredible. Such a great idea to build such sophisticated and polished apps for internal use.
English
0
0
1
8
Brian Casel
Brian Casel@CasJam·
Latest project: SparkDrop — A content pipeline I built for me and my agents to run together. My agent pitches ideas, drafts content, and surfaces learnings every morning at 6am. I review, give input, approve and schedule. 2 weeks in Claude Code. Using it every day.
English
11
3
79
9.5K
Shaan Puri
Shaan Puri@ShaanVP·
I'm secretly launching a 2nd youtube channel reply "send it" and I'll dm it to you
English
918
4
321
103.7K
Andreas Keller retweeté
Claude
Claude@claudeai·
Computer use is now in Claude Code. Claude can open your apps, click through your UI, and test what it built, right from the CLI. Now in research preview on Pro and Max plans.
English
2.6K
4.8K
59.2K
16M
Andreas Keller
Andreas Keller@itsakeller·
@VinodSharma10x I also mostly use Claude Code for development. Really like the speed and quality of the output.
English
0
0
2
31
Vinod Sharma
Vinod Sharma@VinodSharma10x·
46% of developers love one AI tool. The Pragmatic Engineer just published their 2026 AI tooling survey. Claude Code is the most-loved developer tool, at 46%. Cursor sits at 19%. GitHub Copilot at 9%. I use Claude Code every day to build Sucana. So the numbers don't surprise me. But one stat did: 75% of small startups use Claude Code. At large enterprises, GitHub Copilot still dominates. That gap tells you everything about where software is heading. Small teams pick the tool that gives them the most leverage. Big companies pick the tool that fits their compliance checklist. The real shift? 55% of developers now regularly use AI agents. Not copilots. Agents. Tools that inspect your repo, generate changes, run tests, and ship a pull request. We went from "AI helps me write code" to "AI ships features while I review." I build with AI agents every day. Not because it's trendy. Because three founders in three countries can't build a SaaS product without them. The solo founder with AI agents will compete with 50-person teams. That future is already here. PS: What AI dev tool do you use most? And has it changed in the last 6 months?
Vinod Sharma tweet media
English
8
1
11
474
Andreas Keller retweeté
Andrej Karpathy
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.
English
1.6K
4.7K
37.2K
5.1M
Andreas Keller retweeté
Claude
Claude@claudeai·
Ads are coming to AI. But not to Claude. Keep thinking.
English
1.7K
4K
50.9K
5.2M
Andreas Keller retweeté
Guillermo Rauch
Guillermo Rauch@rauchg·
We're encapsulating all our knowledge of @reactjs & @nextjs frontend optimization into a set of reusable skills for agents. This is a 10+ years of experience from the likes of @shuding, distilled for the benefit of every Ralph
Guillermo Rauch tweet media
English
348
602
8.8K
1M
Andreas Keller retweeté
Nathan Lands
Nathan Lands@NathanLands·
I'M BLOWN AWAY. Andrej Karpathy just explained Software 3.0 at YC. BIG IDEAS: English is coding. AI is electricity. And, build for LLMs, not just people. Key takeaways:
Nathan Lands tweet media
English
96
630
6.6K
1.4M
MA
MA@AryanMayya·
@shl Hey @shl not able to access this list :(
English
1
0
2
305
Andreas Keller
Andreas Keller@itsakeller·
5. Be consistent and find ways to monetize the site Adding new content, writing tweets and newsletters and improving the user experience takes consistent effort. To make this last, monetization is key. With more traffic affiliate links and sponsorships should work.
English
2
0
1
105