May 18
Rust live coding - 1:45pm PT - in 25 minutes
Building zcoder, E04
➜ Refine Event Model, fix and code review
➜ Data Store v1, Data Store vs. TUI State
youtube.com/live/P4Ah_YI7QyM
@rustlang #rust @ratatui_rs #ProductionCoding #LiveCoding
5
185
May 17
In today's live stream, we will focus on the first 10% and show the rest as well.
youtube.com/live/rC6sH5DMlbs
@rustlang #AICoding #ProductionCoding #LiveCoding
NO #VibeCoding
1
2
148
Apr 23
Pretty sweet the ew #Zed worktree / git branch top bar UI
Key for parallel AI
@zeddotdev @rustlang #AICoding #ProductionCoding
3
236
Apr 21
My answer to @mattpocockuk
"What 'advanced' AI coding techniques are you using?"
For me, it's #Workbench
vvv VIDEO EXPLAINER vvv
#AICoding #ProductionCoding @rustlang
(not #vibecoding)
2
195
Apr 10
pro@coder v0.5.2, BIG Release
- workbench (dev chat, spec, plan)
- sub/team agents. Event-based (on: ["start", "end"])
- single plan file
- spec support (with spec-rules)
substack.com/profile/6219267…
(#PRD = wrong loop, Be #SpecDriven)
#ProductionCoding #AICoding @AIPackAI
2
4
240
Mar 30
AI is amazing for speeding up code generation.
But that applies to good code and bad code.
The trick ➜ keep flattening the system design complexity curve,
✅ so it stays focused on good code generation
❌ not bad code proliferation
#ProductionCoding #AICoding
2
115
Mar 23
My workflow for AI #ProductionCoding (#rustlang)
➜ Control first, magic second
➜ #workbench & #spec driven AI dev
➜ Everything files, even prompt
➜ Flatten the complexity ⇒ scale code
Soon .. More videos on #workbench dev
CC #mattpocockuk
youtube.com/watch?v=SuLHINfq…
6
391
Mar 20
Hot take on Ralph Loop for Production Coding
I wish developers would focus more on spec-driven development rather than PRD-driven loops.
We might be focusing on the wrong loop.
#AICoding #ProductionCoding #rustlang CC @mattpocockuk
6
215
Mar 19
Vibe coding is awesome for ideation and POCs!
Vibe coding == Figma
And 90% of production coding should be generated by AI for new projects and products.
Engineering is being reshaped, but it is still engineering.
#AICoding #ProductionCoding #VibeCoding
5
4
205
Mar 19
We tested GPT 5.4 Mini vs. Gemini Flash 3
➜ 3x cheaper,
➜ 3x faster
Why? Flash 3 tends to overthink here.
But Flash 3 is, or was, so good.
Those small but good models are awesome for
➜ code maps
➜ doc maps
➜ auto-context
➜ classification, ..
#genai #ProductionCoding
2
3
168
Mar 16
Code Map, 41 files, 32 concurrency, 10 seconds (3 min of AI work)
Code Lensing ➜ Code Map ➜ Auto Context ➜ AI Coding
Quality in, quality out. Cheaper, faster, fewer round trips
#AICoding #ProductionCoding
4
272
Mar 11
Web Dev at Rust speed.
rollup to #rolldown ➜ 30x faster (rolldown.rs)
pcss to #lightningcss ➜ 100x faster (lightningcss.dev)
Now, repl is just a watch & rebuild. No ninja tricks.
#rustlang @WebDevelopment #AICoding #ProductionCoding
2
173
Mar 8
Interesting @OpenAI gpt 5.3 codex was working great 2 days ago, but since yesterday, quality is way down, not sure why.
Back to Opus, it works great.
All via API, so free as in freedom
#AICoding #ProductionCoding
2
254
Mar 7
My new coding pair (via API)
Lite 3.1 - Code Map
Flash - Auto Context
Codex 5.3 - Coding
Somehow, Codex 5.3 is VERY efficient with output tokens, making it so much cheaper than Opus or Sonnet
#AICoding #ProductionCoding #rustlang
3
382
3 Dec 2025
Gemini Flash is the most underrated coding model.
Surprisingly good, decently fast, huge context window, and very cheap.
Pair it with another "big coder" model and you are all set.
Can't wait for gemini-3-flash
@GoogleAI @OfficialLoganK #AICoding #ProductionCoding
1
3
268
30 Nov 2025
Big release of #AIPack 0.8.10 and pro@coder 0.2.28
With attachment support (image and pdf)
See demo below, turning static image to interactive chart.
#vibecoding to #productioncoding @AIPackAI
news.aipack.ai/p/procoder-st…
1
4
237
23 Nov 2025
This seems obvious, but unfortunately some devs still skip basic best practices and Git discipline (especially when using AI).
We do not have to be Git experts to use Git well.
#ProductionCoding #programming
howtogeek.com/beginning-git-…
1
5
290
20 Nov 2025
For anyone interested in a quick pricing comparison:
My current production coding models are:
- Bigger work ➜ Gemini 3 (backup 5.1-codex)
- Simpler work ➜ 5.1-codex-mini (backup flash-latest)
#AICoding #ProductionCoding #Gemini3 @openai
3
1
5
503
7 Oct 2025
👇My new coding pair👇
gpt-5-codex AND gemini-flash-latest
Why:
➜ gpt-5-codex
Slow, but the best coder.
➜ gemini-flash-latest (09-2025)
Fast, cheap, and pretty good!!
Why not #Sonnet 4.5:
➜ doesn't listen well and is expensive.
#genai #AICoding #ProductionCoding
1
3
201
5 Oct 2025
It's sad that #gpt5 has become so slow for #coding (via API).
3x to 5x slower (gpt-5, -mini, -codex)
#Alternatives:
gpt-5-mini ➜ gemini-flash-latest
gpt-5-codex ➜ ~ #Sonnet 4.5
@OpenAI #GiveSomeCodingSpeed #Please
#genai #AICoding #ProductionCoding @AnthropicAI @GoogleAI
1
4
307