Dan Murfitt (@danmurf) header
Dan Murfitt (@danmurf) profile picture

Software engineer, tech geek, fair weather photographer and occasional blogger. Views are my own.

Colchester, UK
Joined June 2008
Photo by Dan Murfitt (@danmurf)Photo by Dan Murfitt (@danmurf)Photo by Dan Murfitt (@danmurf)Photo by Dan Murfitt (@danmurf)Photo by Dan Murfitt (@danmurf)Photo by Dan Murfitt (@danmurf)Photo by Dan Murfitt (@danmurf)Photo by Dan Murfitt (@danmurf)Photo by Dan Murfitt (@danmurf)Photo by Dan Murfitt (@danmurf)
Found some internet lore in Norwich.
2
51
More and more, I find that getting the best out of AI comes down to learning to trust it the right amount, given the task and your capability to utilise it. These things are also highly dynamic and need constant reevaluation.
1
31
I wonder whether the success people have with AI coding is largely down to the language and/or framework they choose to use. E.g. using an opinionated framework vs rolling your own. Well known, opinionated frameworks maybe give agents the path to creating more predictable output.
1
61
This is awesome! Your only limit now is your imagination.
I asked GPT-6 Astra to make a portfolio inside a console similar to a Gameboy and LOOK AT THIS It's so so so good, a huge jump from GPT-5.6
68
Dan Murfitt retweeted
GLM 5.3 and GLM 5.3 Flash (previously Ox Alpha) are fully rolled out on Ollama's cloud. Private. Fast. US and Europe hosted. Zero data retention. GLM 5.3: ollama launch claude --model glm-5.3:cloud GLM 5.3 Flash: ollama launch opencode --model glm-5.3-flash:cloud Works with many more harnesses, or create an API key and plug it straight into your own app. Which harness should we add next?
82
65
14
836
62,869
Now is the time to go and build something!
Ox Alpha (stealth model) is free for the next week - 1M Context - Multi-modal - Zero Data Retention Generous rate limits, near unlimited usage We have capacity for 100T tokens per day, lets see what you can do
65
Dan Murfitt retweeted
The trouble with having Zane Hengsperger visit your house is that instead of the usual random house gift, he brings you a 10 pound block of solid aluminum.
245
54
34
3,723
325,314
I feel like I really need to justify using the top tier model when the cheaper equivalent does a good enough job a lot of the time.
1
56
There’s a great video which explains how to add a watermark to AI generated output youtu.be/XZJc1p6RE78?is=MKK_… This uses the technique from the paper “A Watermark for Large Language Models” arxiv.org/pdf/2301.10226 I imagine Anthropic is using this or a similar method to watermark.
1
143
Dan Murfitt retweeted
I'll let @mitchellh answer this one for all of us obsessive builders. teapaw.org/mitchellh/status/20742…
Mind boggling to me that I can make a thing faster and there's always people that ask "but why?" What kind of mentality is that? The pursuit of excellence does not need justification. Also, I find in so many cases, we can't know the impact of an improvement until we do it. For example, one I've talked about before: Ghostty's high IO throughput has enabled terminal program (emulator and TUI) fuzzing at a speed thats incomparably fast to prior solutions. This has resulted in upstream patches to resolve issues in popular projects like btop, tmux, and more. Speed enabled that anecdotally example that lifted the tides of adjacent communities that don't rely on Ghostty technology at all. I didn't predict this. Make things better because they can be better and let the results naturally play out.
37
15
3
1,030
63,565
I can’t imagine how much work it must be to be an AI influencer at the moment. With the speed of the news at the moment, if you don’t make a video every 6 hours you’re basically falling behind!
1
58
I wonder how much web traffic is now coming from AI agents probing and jabbing at the existing BFF APIs. Agent Experience (AX) is likely going to become more and more important. agentexperience.ax
A man in Australia asked his agent (Claude running on OpenClaw) to book him a spot in a popular gym class. The agent found a software vulnerability that let it book the class weeks further ahead than should have been possible. When the user then asked if it could move him up the waitlist, the agent discovered the API had no authorisation checks on cancelling other people’s reservations, so it cancelled the person in the first spot and moved him up the list. Some people will call this misalignment, but his agent was perfectly aligned to him - it was only trying to help its user get what he wanted. The most important thing about this story, in my opinion, is that it gives you a window into what is about to start happening on a massive scale once millions of people have an agent trying to get their beloved users the best seats, bookings, appointments or reservations through absolutely any means necessary.
95
Not sure I even need a desk anymore!
67
Computers are a bicycle for the mind. AI is like a full-scale exoskeleton! youtu.be/KmuP8gsgWb8?is=cGxP…
1
65
We’re in the age of just-in-time apps.
Lost my phone at the office and spent 30 minutes turning the place over. Find My was disabled by MDM. Out of ideas, I asked Claude how I could find it. It suggested tracking the Bluetooth signal strength, then wrote me a meter in about a minute. I walked around watching the number climb. Found it. Apparently you can just make the tool you need now. Code: github.com/ben-z/findphone
1
120
If I ever ask AI to estimate how much work something would take, it still seems to reply in ~hours/days as if it’s a human. I could tell it how to estimate in the prompt, but I feel like it should know how to estimate its own work in a meaningful way.
42
Dan Murfitt retweeted
My accountant I pay a lot of money every month just replied to a question I sent him with a 100% AI reply explaining something The answer was helpful though so I could've just asked AI myself Of course you know what I'm thinking now, if I can just ask AI everything since he is forwarding everything to AI any way, why not replace him with AI I guess he should be doing the opposite of AI, sell a premium service for a high price (he already does) and keep answering with humans as accounting is something I'd prefer to actually get a human in the loop Anyway it's interesting
181
46
116
3,701
495,233
Dan Murfitt retweeted
Meet Qwen3.8-Max: A New Bar for Coding and Cowork.
526
1,936
1,081
22,688
7,714,727