Rory MacLeod

94 posts

Rory MacLeod

Rory MacLeod

@rjmacleod_dev

evals n chill

Mountain View, CA Katılım Aralık 2025
223 Takip Edilen6 Takipçiler
Rory MacLeod
Rory MacLeod@rjmacleod_dev·
My early 2024 IDE: VScode + copy paste code snippets from ChatGPT Late 2024 - Mid 2025: Cursor Late 2025: Claude Code + VScode Early 2026: Whispr flow -> telegram -> Claude Code on mobile, ghostty + subdivided windows of Claude code agents on desktop Crazy how fast things are changing
English
0
0
0
13
Rory MacLeod
Rory MacLeod@rjmacleod_dev·
Another day, another attempt to use @claudeai /remote-control only to have it lose connection in 5 minutes They keep shipping new features which makes me wonder if last week’s features were fixed, but sadly no.
English
0
0
0
6
hope hopes hoping
hope hopes hoping@hopes_revenge·
Claude .md ? he’s a doctor too?
English
38
230
4.7K
105.2K
Rory MacLeod
Rory MacLeod@rjmacleod_dev·
@dejavucoder I’m convinced the people posting about it aren’t actually using it for more than 5 minutes
English
0
0
2
200
sankalp
sankalp@dejavucoder·
bruh claude code remote control session does not connect half the time. i thought my wifi had issues.
English
10
1
56
11.3K
Rory MacLeod
Rory MacLeod@rjmacleod_dev·
@natolambert Agreed, Opus 4.6 for building the spec and pair work, 5.4 for implementing and long running agentic tasks..
English
0
0
0
272
Nathan Lambert
Nathan Lambert@natolambert·
GPT 5.4 didn't get enough praise for how big of a step it was in OpenAI's agent arc. At the same time, with better context management, speed, rate limits, instruction following, code -- it's revealing that I still turn to the "warmth" of Claude. interconnects.ai/p/gpt-54-is-a-…
English
22
22
302
27K
Rory MacLeod
Rory MacLeod@rjmacleod_dev·
@GergelyOrosz “Probably not for long tho” 💯 I’ve realized I appreciate seeing mistakes in typing and clues that I’m reading human generated text.. I’m sure the bots will catch on to this soon
English
0
0
0
74
Gergely Orosz
Gergely Orosz@GergelyOrosz·
It’s not X — it’s Y I cannot unsee how so much of the writing on this site (and online, in general) is increasingly AI-generated. It’s still pretty easy to recognize. Probably not for long tho Just alarming that ppl outsource even typing 3 sentences for a reply on this site…
English
154
33
1.2K
45.6K
Rory MacLeod
Rory MacLeod@rjmacleod_dev·
The @claudeai opus 4.6 1M context length in Claude code is amazing, so much bigger than 4.5 -> 4.6
English
0
0
0
25
Rory MacLeod
Rory MacLeod@rjmacleod_dev·
Feels like someone smarter than me could make an analogy to classical / quantum physics (for classical software dev vs. Agentic engineering). There’s an inherent uncertainty the finer down you drill into the state of the code. Unfortunately we don’t get the reproducible probabilistic outcomes from agents.
English
0
0
0
160
Hamel Husain
Hamel Husain@HamelHusain·
One thing that makes me feel that code factory has not arrived yet is the following experiment: 1.Ask a LLM to do an in-depth rigorous review of your code 2. In a new thread, as same/different LLM to consider those review comments independently and address issues it agrees with 3. Keep repeating until no new concerns I find that this loop always goes on for a ridiculously long time, which means that there is a problem with the notion of claude-take-the-wheel. This seems to happen no matter the harness or the specificity of the specs. It works fine for simple applications, but in the limit if the LLMs have this much cognitive dissonance you cannot trust it. Either this, or LLM are RLHFd to always find some kind of issue.
English
69
10
251
26.8K
Rory MacLeod
Rory MacLeod@rjmacleod_dev·
@noahzweben Remote control is such a tease. It works intermittently between my Mac mini and iPhone, and lacks functionality from the cli. I usually get 1 or 2 messages in then need to jump back into Termius/tmux to get things unstuck again
English
1
0
2
78
Rory MacLeod
Rory MacLeod@rjmacleod_dev·
@trq212 It mainly seems to trigger when I get upset about something. Makes me feel like the feature is some form of PTSD. I want it to remember the good times, too
English
0
0
0
369
Rory MacLeod
Rory MacLeod@rjmacleod_dev·
@Steve8708 I pay for Claude to review all my vibe coded PRs just for the legal liability - Anthropic’s got deeper pockets than me
English
0
0
0
500
Steve (Builder.io)
Steve (Builder.io)@Steve8708·
anthropic literally charged us $25 for this
Steve (Builder.io) tweet media
English
119
41
1.4K
264K
Rory MacLeod
Rory MacLeod@rjmacleod_dev·
@testinprodcap Jus stick a WMD on a UAP.. some C4 on a FPV..maybe even a BFG on a H1B…
English
0
0
0
11
Rory MacLeod
Rory MacLeod@rjmacleod_dev·
@TheZvi Did OpenAI do this so they would be better strategically positioned for fast takeoff? I don’t quite have a model for how this would work, but I’m trying to understand why they were so eager for the contract.
English
1
0
1
173
Zvi Mowshowitz
Zvi Mowshowitz@TheZvi·
Open ended: What would be the most valuable short thing for me to write about today re: Anthropic/DoW/OAI, if any? Since nothing new is on fire, which is great.
English
17
0
41
4.4K
Rory MacLeod
Rory MacLeod@rjmacleod_dev·
@steipete My 8gb m2 Mac mini works pretty well. It’s a good tradeoff to get it to $599
English
0
0
1
1.5K
Rory MacLeod
Rory MacLeod@rjmacleod_dev·
@signulll @Noahpinion Claude Team being so much bigger than Claude Enterprise is shocking to me. Who is using Claude Team?
English
0
0
1
394
signüll
signüll@signulll·
this is quite a ridiculous chart if accurate.
signüll tweet media
English
98
203
3.6K
263.3K