Hacker News: cleiferhttps://news.ycombinator.com/user?id=cleiferHacker News RSShttps://hnrss.org/hnrss v2.1.1Thu, 30 Jul 2026 09:11:28 +0000<![CDATA[Clownmaxxing, experimental prompts and Joey's cadillac todo list]]>Article URL: https://charlesleifer.com/blog/clownmaxxing-experimental-prompts-and-joey-s-cadillac-todo-list/

Comments URL: https://news.ycombinator.com/item?id=48598242

Points: 4

# Comments: 0

]]>Fri, 19 Jun 2026 13:18:15 +0000https://charlesleifer.com/blog/clownmaxxing-experimental-prompts-and-joey-s-cadillac-todo-list/cleiferhttps://news.ycombinator.com/item?id=48598242https://news.ycombinator.com/item?id=48598242<![CDATA[Tokens and Dreams]]>Article URL: https://charlesleifer.com/blog/tokens-and-dreams/

Comments URL: https://news.ycombinator.com/item?id=48561369

Points: 2

# Comments: 0

]]>Tue, 16 Jun 2026 20:14:47 +0000https://charlesleifer.com/blog/tokens-and-dreams/cleiferhttps://news.ycombinator.com/item?id=48561369https://news.ycombinator.com/item?id=48561369<![CDATA[Cave of Forgotten Dreams]]>Article URL: https://charlesleifer.com/blog/cave-of-forgotten-dreams/

Comments URL: https://news.ycombinator.com/item?id=48434582

Points: 4

# Comments: 0

]]>Sun, 07 Jun 2026 13:23:14 +0000https://charlesleifer.com/blog/cave-of-forgotten-dreams/cleiferhttps://news.ycombinator.com/item?id=48434582https://news.ycombinator.com/item?id=48434582<![CDATA[New comment by cleifer in "Mythical Man Month"]]>Loved Brooks’ and feel like I must’ve internalized his model because that’s all I care about anymore in dev. Coherence among the components (people or software) to produce something good. I think of if 100 pianos are tuned to the same tuning fork then they all play well together.

AI accelerated iteration and undisciplined use seem to me the main risk lately. Drift from any prompting, accumulation of features, plausible looking patches that are actually slop. Wrote about this recently [0], it creates a dynamic where you aren’t just making the software less coherent, you’re paying more and more to make it so. Curious to hear others experience of this.

[0] https://charlesleifer.com/blog/tokens-and-dreams/

]]>Sat, 09 May 2026 13:20:21 +0000https://news.ycombinator.com/item?id=48074765cleiferhttps://news.ycombinator.com/item?id=48074765https://news.ycombinator.com/item?id=48074765<![CDATA[Tokens and Dreams]]>Article URL: https://charlesleifer.com/blog/tokens-and-dreams/

Comments URL: https://news.ycombinator.com/item?id=48025750

Points: 2

# Comments: 0

]]>Tue, 05 May 2026 17:32:23 +0000https://charlesleifer.com/blog/tokens-and-dreams/cleiferhttps://news.ycombinator.com/item?id=48025750https://news.ycombinator.com/item?id=48025750<![CDATA[New comment by cleifer in "Claude Code daily benchmarks for degradation tracking"]]>How much influence have you all found prompting to have on output quality? Generally I've been approaching by just describing my problem and assuming that I'll get the machine's optimal output, but perhaps being explicit in the prompt can impact the output quality?

]]>Fri, 30 Jan 2026 14:39:54 +0000https://news.ycombinator.com/item?id=46824941cleiferhttps://news.ycombinator.com/item?id=46824941https://news.ycombinator.com/item?id=46824941<![CDATA[Ghost in the Shell: My AI Experiment]]>Article URL: https://charlesleifer.com/blog/ghost-in-the-shell-my-ai-experiment/

Comments URL: https://news.ycombinator.com/item?id=46824742

Points: 2

# Comments: 0

]]>Fri, 30 Jan 2026 14:24:41 +0000https://charlesleifer.com/blog/ghost-in-the-shell-my-ai-experiment/cleiferhttps://news.ycombinator.com/item?id=46824742https://news.ycombinator.com/item?id=46824742