Iarfhlaith Kelly

10.7K posts

Iarfhlaith Kelly banner
Iarfhlaith Kelly

Iarfhlaith Kelly

@Iarfhlaith

CTO. Former founder. Tweets mostly about the business of software.

Dublin, Ireland. Katılım Mart 2008
815 Takip Edilen1.3K Takipçiler
Iarfhlaith Kelly
Iarfhlaith Kelly@Iarfhlaith·
Really enjoyed watching Flight 12 last night. Congrats @SpaceX. The big question now is whether this heat shield could go straight back up again with zero refurb.
Iarfhlaith Kelly tweet media
English
0
0
0
38
Tibo
Tibo@thsottiaux·
@Mugilan_SS @OpenAI It’s the small things. When we designed this the tradeoff was between a good experience or optimize for margin and not allow to draw more usage than you technically have. We chose to optimize for the experience as it’s really annoying to have your agent interrupted midway.
English
123
66
3.5K
146K
Mugilan S
Mugilan S@Mugilan_SS·
Codex is not like claude code. if you know the limit is going to end, like last 10 to 8%, give an very long run task, and even after the limit got ever, it will continue to do the task until the task was completed. Shout out to @OpenAI team.
English
308
399
13.1K
1.4M
Santiago
Santiago@svpino·
How many of you think that “software developer” will be a viable career in 3-5 years?
English
224
8
368
145.2K
Iarfhlaith Kelly
Iarfhlaith Kelly@Iarfhlaith·
This is bananas. CEOs quite literally begging Dario not to render their companies irrelevant. In fairness, Anthropic have, so far, focused mostly on enabling the creators rather than replacing the business apps they create. But, that might come too.
Bojan Tunguz@tunguz

English
0
0
0
66
Iarfhlaith Kelly
Iarfhlaith Kelly@Iarfhlaith·
@gsivulka OpenAI, Anthropic, xAI. The companies that enabled the productivity captured that value.
English
0
0
0
304
Iarfhlaith Kelly
Iarfhlaith Kelly@Iarfhlaith·
@chinmay185 The limiting factor in rolling this out across a medium to large org is management’s capacity to facilitate and support the change.
English
0
0
1
54
Chinmay Naik
Chinmay Naik@chinmay185·
I was chatting with someone who has integrated AI into their SDLC fully. Basically, to create stories from high-level requirements, using sub-agents and skills to generate an implementation plan, architecture, code, tests, deployment pipelines, etc. When I talk to folks using this kind of workflow, here's something I am seeing. The bottleneck has shifted to how fast you can execute your test suite, how "confident" you are in code and the tests and what kind of observability and guardrails you have on production to be able to find and fix issues quickly. Interesting times. Folks who have been using agentic-engineering workflows, what are your observations/thoughts? What are some bottlenecks are you seeing?
English
11
1
33
5.3K
Tibo
Tibo@thsottiaux·
With GPT-5.4 out. What should Codex ship or improve next?
English
1K
16
1.2K
113.1K
Iarfhlaith Kelly
Iarfhlaith Kelly@Iarfhlaith·
LOL
malinvestment.jpeg@malinvested

Of course that's your contention. You're a first-time SaaS bear. You just got finished listening to some podcast, Dario on Dwarkesh, probably. Now you think it’s the end of white collar work and seat-based pricing is screwed. You're gonna be convinced of that til tomorrow when you get to “Something Big is Happening”. Then you’ll install ClawdBot on a Mac Mini, vibe code a dashboard on top of a postgres database and say we’re all just a couple ralph loops away from building a Salesforce competitor. That’s gonna last until next week when you discover context graphs, and then you're gonna be talking about how the systems of record will be disintermediated by an agentic layer and reposting OAI marketing graphics. “Well, as a matter of fact, I won't, because ultimately the application layer is just ….” The application layer is just business logic on top a CRUD database. You got that from Satya’s appearance on the BG2 pod, December 2024, right? Yeah, I saw that too. Were you gonna plagiarize the whole thing for us? Do you have any thoughts of your own on this matter? Or...is that your thing? You get into the replies of anyone posting a SaaS ticker. You watch some podcast and then pawn it off as your own idea just to impress some VCs and embarrass some anon who’s long SaaS? See the sad thing about a guy like you is in a couple years you're gonna start doing some thinking on your own and you're gonna come up with the fact that there are two certainties in life. One: don't do that. And two: you dropped thirty grand on Mac Minis and LLM API calls to come to the same conclusion you could’ve got for free by following a handful of VC accounts.

QST
0
0
0
96
Iarfhlaith Kelly
Iarfhlaith Kelly@Iarfhlaith·
Read this. I would add that AI is disrupting coding first because 1. coding infrastructure already has version control and code review practices effectively providing HITL capabilities already; and 2. a community built on embracing change and adopting the state of the art.
Matt Shumer@mattshumer_

x.com/i/article/2021…

English
0
0
0
150
Iarfhlaith Kelly
Iarfhlaith Kelly@Iarfhlaith·
@_maxq_ @NASASpaceflight Flight 8’s gap between ship static fire and launch was 25 days—ship static fire was a Feb 11th, launch was Mar 6th. So we can roughly expect launch of Flight 9 to be last week in May. 🚀
English
1
0
0
238
World of Statistics
World of Statistics@stats_feed·
Year founded: 🇮🇹 University of Bologna: 1088 🇬🇧 Oxford: 1096 🇬🇧 Cambridge: 1209 🇺🇸 Harvard: 1636 🇨🇦 University of Toronto: 1827 🇨🇭 ETH Zurich: 1855 🇺🇸 Massachusetts Institute of Tech: 1861 🇺🇸 California Institute of Tech: 1891 🇺🇸 Stanford: 1891 🇬🇧 Imperial College London: 1907
English
128
186
1.7K
176K
Iarfhlaith Kelly
Iarfhlaith Kelly@Iarfhlaith·
@Padday Agree, but would extend to all written words. When the cost of writing goes to zero, proof of (human) work gives the material a lot more credibility.
English
0
0
1
76
Paul Adams
Paul Adams@Padday·
Hot take: Marketing that 'feels like marketing' will start to perform badly. Marketing will need to feel authentic, raw, and real to be credible. We're entering an era where AI can make almost anything, and so people will place more value on things that are made by humans. Real words. Spoken by real people. Spoken like you'd speak to a friend or colleague. No buzzwords. No catchphrases. No clever wordplay.
English
4
0
15
1.8K