Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
tomjohnneill
searching Neon…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
9 ms
·
1.
▲
by
tomjohnneill
1y ago
Did the agents you built work? I think we're doing something similar in the UK and intrigued to know how it went for you ( https://www.plinth.org.uk )
2.
▲
by
tomjohnneill
1y ago
I'd be very interested to hear more about how the non-for-profit agents platform went/anything else you learnt.
3.
▲
Major breakthroughs in UK munitions production
(baesystems.com)
46 points
by
tomjohnneill
1y ago
|
40 comments
4.
▲
by
tomjohnneill
2y ago
And it's ASML more than TSMC. Or maybe there's a highly profitable role for all the different parts of the value chain.
5.
▲
by
tomjohnneill
2y ago
Plinth ( https://plinth.org.uk/ ) | London (UK) | Senior Engineer (mostly Frontend) | Onsite At plinth, we've built a software/AI platform for charities to help them measure their impact and get more funding for the
6.
▲
by
tomjohnneill
3y ago
I think it's a case of tech bubble vs the rest of the world. Most people are not subscribing to the paid version of ChatGPT, but a lot of people who spend a lot of time with these things are.
7.
▲
by
tomjohnneill
3y ago
It's the default for the free version of ChatGPT no? That's what the majority of people use.
8.
▲
by
tomjohnneill
3y ago
I'm not sure. The figures I've seen suggest that GPT3 required 10x more energy to train than GPT2 (e.g. https://www.nnlabs.org/power-requirements-of-large-language-... .), so I think a roughly 1-2 order of magnitud
9.
▲
by
tomjohnneill
3y ago
I can definitely imagine they're not covering the amortised cost of the training with the cost per individual inference request. It seems less likely to me that they're making a significant loss on each subsequent request, but aga
10.
▲
by
tomjohnneill
3y ago
That's fair about ignoring the training cost. I did write a bit more going into that in a follow up piece here: https://notfunatparties.substack.com/p/ai-is-good-for-the-pl... Do you have any better sources for th
11.
▲
Paragraph Pollution: AI is (probably) greener than you typing on a laptop
(blog.plinth.org.uk)
3 points
by
tomjohnneill
3y ago
|
12 comments
12.
▲
Cocytarchy – By Kulak – Anarchonomicon
(anarchonomicon.substack.com)
2 points
by
tomjohnneill
4y ago
|
0 comments
13.
▲
by
tomjohnneill
4y ago
Maybe decoding Herculaneum scrolls? https://en.wikipedia.org/wiki/Herculaneum_papyri#Virtual_unr...
14.
▲
by
tomjohnneill
4y ago
It's been done (unsurprisingly): https://www.theaccessgroup.com/en-gb/products/mosaic/
15.
▲
by
tomjohnneill
4y ago
That's actually not correct. Stripe invoicing charges 0.4 or 0.5% on top of the normal payment transaction fee (for either card payments or bank transfers). Essentially, you are paying a % fee for a nice PDF (plus some reconciliation&#
16.
▲
by
tomjohnneill
4y ago
They fixed that fairly recently.
17.
▲
by
tomjohnneill
5y ago
I'm not sure the API docs are publicly accessible?
18.
▲
by
tomjohnneill
5y ago
It's very likely organisations in the charity sector will need to run DBS checks, even just for looking at back-end systems/data.
19.
▲
Optimum Wordle Solver
(notfunatparties.substack.com)
1 points
by
tomjohnneill
5y ago
|
0 comments
20.
▲
by
tomjohnneill
5y ago
Interesting to read those valuation benchmarks. I'd seen this article in SeedLegals previously https://seedlegals.com/resources/how-to-value-your-company/ that seem to be quite a lot lower. It was written at
21.
▲
by
tomjohnneill
5y ago
Paywalled, but: https://theathletic.com/2755968/2021/08/09/cox-is-heading-ab... The main suggestion is that heading could be banned outside the 2 penalty boxes.
22.
▲
by
tomjohnneill
5y ago
A UBI that doesn't apply to foreign migrants isn't exactly universal. Also, the second order impact of oil money appreciating the currency and therefore the reducing the international competitiveness of non-oil industries definite
23.
▲
by
tomjohnneill
5y ago
Yeah it was statically hosted in the end (initially it wasn't quite static). When it was at its most popular, the total site size including JavaScript was around 70kb as far as I can remember, so it didn't come anywhere near as hi
24.
▲
by
tomjohnneill
5y ago
Part of my worry was that I messed up the caching initially so it was making API calls on the server for each request. Once I fixed that, there wasn't really anything to worry about.
25.
▲
by
tomjohnneill
5y ago
It was basically a typo, but then enough people got slightly irritated by it that I started to enjoy having it that way.
26.
▲
by
tomjohnneill
5y ago
It was a static site cached on Vercel. Not including the map (which is an iframe) the site was less than 100kb.
27.
▲
by
tomjohnneill
5y ago
I wrote that up here: https://notfunatparties.substack.com/p/inside-a-viral-websit...
28.
▲
by
tomjohnneill
5y ago
Provided you gain appropriate consent, there are definitely ways in which you can discover exactly how much the taste of these fluids vary from person to person.
29.
▲
by
tomjohnneill
5y ago
Yeah I don't think anyone is ever going to build a better Goodreads from a bookmarking & automated recommendations perspective for exactly that reason. I think there is scope for some nice little niche tools for individuals to reco
30.
▲
What Goodreads could have been
(notfunatparties.substack.com)
93 points
by
tomjohnneill
5y ago
|
35 comments
More ›