Paul Ingles (@pingles) header
Paul Ingles (@pingles) profile picture

CTO. happy hacker. https://nitter.gq/t.co/LzQvxhP4mu. Alum: https://nitter.gq/t.co/9QMSq2KgtE, ThoughtWorks. email: https://nitter.gq/t.co/OZjjk6IHGi…

London
Joined March 2008
Photo by Paul Ingles (@pingles)Photo by Paul Ingles (@pingles)Photo by Paul Ingles (@pingles)Photo by Paul Ingles (@pingles)Photo by Paul Ingles (@pingles)Photo by Paul Ingles (@pingles)Photo by Paul Ingles (@pingles)Photo by Paul Ingles (@pingles)Photo by Paul Ingles (@pingles)Photo by Paul Ingles (@pingles)
From Outropy.ai's first investor update: using LLMs is racing against the Fail Whale.
2
1
972
Imagine looking at this chart and thinking "yep this looks legit". But naturally, despite it being glaringly wrong and misleading, and despite this having been pointed out to the author, it remains up.
59
260
95
1,791
540,777
AI now getting high marks on standardised exams is impressive, but I think it also says a lot about how many exams are a terrible way of assessing human knowledge... 1/
8
21
2
130
25,132
Scrum became ubiquitous because it answered two questions at once: "What does good look like" -> following Scrum "How do we get to good" -> follow Scrum We need to give product leaders looking for a definition of good for their team a better answer than "following Scrum."
5
1
32
3,608
XP never had a story about scaling, which limited its reach. I get the feeling that there is a good story to be told about helping 1000s of folks coordinate their efforts & that story starts by reducing team dependencies. tidyfirst.substack.com/p/sca…
2
30
3
100
21,400
Those meddling kids
2
54
4
237
learning to code in R
161
1,206
335
11,239
Paul Ingles retweeted
Falling short of a goal doesn't mean failure. Often it's progress. 14 experiments, 10k people: people counted results that didn't meet targets as losses—even if they were gains. The key measure of success isn't reaching your destination. It's improving from your starting point.
24
523
72
2,222
Paul Ingles retweeted
Of all of the “dangers of AI” papers, this is most worrying: AI researchers building a tool to find new drugs to save lives realized it could do the opposite, generating new chemical warfare agents. Within 6 hours it invented deadly VX… and worse things nature.com/articles/s42256-0…
151
2,192
550
8,870
Paul Ingles retweeted
An underrated reason why some ambitious high slope people change jobs every few years is that their org/company will always maintain some residual memory of their skills & abilities from 1-2 yrs ago and cannot fully appreciate who they are now. A job change resets the y-intercept
43
230
33
2,467
A study of compound swears used in Reddit comments from 2006 to 2020 has found that the most popular combination is ‘dumbass’ (3.6 million comments). The 444 combinations used only once include ‘sleazenozzle’ and ‘bastardbucket’. (📷 and the study: Colin Morris, @HalfEatenScone.)
62
218
110
1,379
Hugh Grant tweeted a request at activists protesting outside Westminster to play the Benny Hill theme on their loudspeakers; when they did it became the soundtrack for street interviews with leading Tories trying explain the situation to the British people
1,905
22,406
8,063
123,755
Paul Ingles retweeted
Last month, @keunwoo published a critical review of Accelerate. @nicolefv and I thought it was a good opportunity to address questions about our work and methods. Here's our response. medium.com/@jezhumble/respon…
5
35
9
137
Paul Ingles retweeted
3
21
1
128
.@BorisJohnson .@10DowningStreet Just touching base - we feel the time is right for a press conference. We have some experience with this sort of thing…
588
8,617
1,907
60,243
Paul Ingles retweeted
Next time you think that the solution to your Infosec problem, lies in an arbitrary list of 16 specialisms you must have to "do it right"... know about the destructive power of handoffs in organisations and what that implies to value creation structures in your organisation
6
14
5
57
Paul Ingles retweeted
Was just reminded of a story that I wrote a detailed blog a few years back about securitydifferently.com/secu… How I saved a big corp over £1m a year, with simple process changes. Big companies are full of inefficiencies like this, your security ones may contribute to that too
1
5
the number of times I have thought "surely there's no way this is made in excel" only to discover that yes, in fact, it is made in excel
131
2,304
272
26,010
Paul Ingles retweeted
How the ‘decoy effect’ influences our decisions from the Predictably Irrational videobook by @litvideobooks litvideobooks.com/irrational
1
12
1
65
Information security research has long established three best practices for websites to help users pick stronger passwords. In a new study, we reverse engineered 120 popular English language websites, and found that only 15 (!) of them follow these guidelines. 🧵
8
116
16
248