ernsheong

47 ms·

1.
by ernsheong 10d ago
They should resurrect the Commonwealth into a stronger union of sorts instead of just playing sports every four years.
2.
by ernsheong 10d ago
Mino 2.5 has been my workhorse for coder and tester agents (the ones planner agents delegate tasks to)
3.
by ernsheong 10d ago
Hey but that's exactly as per design. It is the customer's responsibility to store stuff elsewhere as DR backup, not AWS.
4.
by ernsheong 11d ago
This is potentially huge and can crash the Big Two's stock prices or block their IPOs completely.
5.
by ernsheong 12d ago
Incoming business slop
6.
by ernsheong 13d ago
sounds like you might need to beef up your harness first, and run multi-agent verification loops
7.
by ernsheong 14d ago
Lol the page doesn't even scroll
8.
by ernsheong 14d ago
There's still 3.1 Pro though, as ancient as it sounds now
9.
by ernsheong 14d ago
Flash is the new Pro, try it first
10.
by ernsheong 15d ago
I built my harness in pi within herdr, I cloned (zipped and downloaded) 0xRichardH/pi-herdr-subagents and went from there, and used pi to build itself, adding gate workflow state control, provider fallbacks (I use many token plans), su
11.
by ernsheong 16d ago
Yeah but engineering is still subject to failure modes, AI can't circumvent that short of formal proofing (which nobody really wants to get into)
12.
by ernsheong 16d ago
It seems like Shopify has fallen into a trap thinking that more complexity costs nothing because of AI. In the age of AI we have to embrace the same principle as before that complexity needs to be tamed, not multiplied. Even as humans strug
13.
by ernsheong 16d ago
Wow, citing we need to figure out how to pay out bills and rejecting PR based on that is such a douche move. The fact is Tailwind core is an MIT-licensed open source project.
14.
by ernsheong 17d ago
Toyota was right all along.
15.
by ernsheong 17d ago
"We couldn't continue with that account. Try again." Is it geo-restricted to the US for now?
16.
by ernsheong 21d ago
How are you handling subagents? Subtabs?
17.
by ernsheong 21d ago
Hey there, nice one. Care to share a bit what is under the hood? What are the challenges of local AI? It is RAG-ging? What small models do you use?
18.
by ernsheong 23d ago
Okay, now remove the 2-stone handicap
19.
by ernsheong 24d ago
This is getting tiring.
20.
by ernsheong 26d ago
It's the opposite for me, AI takes away the drudgery, and I want to build everything I can think of.
21.
by ernsheong 27d ago
Thanks for feedback, I've added a horizontal/split view so that images align on one side and titles can be scanned top down.
22.
Show HN: HN Mirror with Image Previews(hackernews.beliantech.com)
3 points by ernsheong 28d ago|2 comments
23.
by ernsheong 28d ago
https://www.madinamerica.com/2021/09/antidepressant-effects-... and its variants
24.
by ernsheong 29d ago
With this chart we now know why stepping down from 15mg to 10mg is a HUGE step.
25.
by ernsheong 29d ago
Basically AI can do the job now. This kind of tool has been nullified by Gemini Notebook, Claude Cowork, etc. or any LLM chat for that matter
26.
by ernsheong 29d ago
Lookup "occupancy chart antidepressant", study the chart. There is an exponential drop-off at lower doses. Hence to taper off without feeling it, one needs to cut smaller and smaller doses, (or liquify) which may be impractical.
27.
by ernsheong 1mo ago
Gosh, instead of wasting time vibing another slop project, stop using Claude Code, use pi or whatever, own the harness, and explore the plethora of models out there. You guys are software engineers, yes?
28.
by ernsheong 1mo ago
There should be a ban on using Open* for for-profit VC-backed companies.
29.
by ernsheong 2mo ago
These people can't version control properly, V4.1 or V5 would be more appropriate.
30.
Show HN: Bananarr – Technology Profiler – Chrome Extension(bananarr.com)
1 points by ernsheong 2mo ago|0 comments

More ›