louiereederson

6 ms·

61.
by louiereederson 7mo ago
TBPN seems like the media equivalent of Soylent. Oh wait...
62.
by louiereederson 7mo ago
Funny how everyone gravitated towards analysis of the star distribution of REPOS when the headline claim is on ADDITIONS. If you look at my comment below (invite you to verify the stats), the distribution of additions by star count is far m
63.
by louiereederson 7mo ago
They announced a $30b+ commitment with Azure back in November, so not really a surprise that Azure is now listed as one of their cloud providers. https://www.anthropic.com/news/microsoft-nvidia-anthropic-an...
64.
by louiereederson 7mo ago
OP. I'm not a confident SQL user so encourage someone to double check this. From what I can tell have been 2.21x10^12 additions on GitHub in total, 6.36x10^11 of which are in repos with 2+ stars. That's about 29%. People earlier w
65.
by louiereederson 7mo ago
That is fair, in fact I just came across their recent blog post on this. They're pointing to usage growth as the issue https://github.blog/news-insights/company-news/addressing-gi...
66.
by louiereederson 7mo ago
Just to clarify as OP, the point here is not that Claude is not contributing to serious work, just that the dashboard suggests a lot of usage in public GitHub repos seems to be tied to low attention, high LOC repos. This is at least somethi
67.
by louiereederson 7mo ago
I hope you don't mind, I thought this was a really valuable dashboard.
68.
by louiereederson 7mo ago
The instability is related to their Azure migration isn't it? Cynically you could say it hasn't been helped by the rolling RIFs at Microsoft
69.
90% of Claude-linked output going to GitHub repos w <2 stars(claudescode.dev)
337 points by louiereederson 7mo ago|222 comments
70.
by louiereederson 7mo ago
Toggling the stars shows 50b lines of code created across all projects, only 5b on projects with 2+ stars since Claude Code launch. Kind of eye opening where these Claude Code tokens are going. Came across this from this ShowHN post yesterd
71.
by louiereederson 7mo ago
This is great, thank you! I've been looking for something like this for a while now. Interesting that whoever is developing BroadwayScore.com appears to be one of the biggest users of Claude Code, if I'm reading this correctly (25
72.
by louiereederson 7mo ago
That would not be stochastic, just parroting
73.
by louiereederson 8mo ago
I think for your comment to follow the guidelines, you need to explain why the original comment did not follow them. Customer values are relevant to the discussion given that they impact choice and therefore competition.
74.
by louiereederson 8mo ago
And they're reportedly back in talks with the DOW per the FT (below). They are not the exception, and are just as bloodlessly, shamelessly publicity hungry as any other tech co, if not more so. No surprise based on their conduct up unt
75.
by louiereederson 8mo ago
Apple is a two sided market between developers and users. OpenAI has not succeeded in building this so far.
76.
by louiereederson 8mo ago
This article is bad. It is mixing up capex and opex. OpenAI is projecting more spending on compute through their income statement now than they were 6 months ago.
77.
CircleCI study: code throughput up, delivery down, reliability slipping(circleci.com)
2 points by louiereederson 8mo ago|1 comments
78.
by louiereederson 8mo ago
Seems like code production is up but mainly in feature branches. Main branch changes are down. MTTR deteriorating. This is for all but the highest performing (top 5%ile) teams. This has to be driven by AI adoption. I'm trying to keep a
79.
by louiereederson 8mo ago
I think this was the best write up on the impact of AI on software engineering I've read yet. By extension it might be the best 'take' on AI I've read period. Alas it is not written by AI and boosted on X (which is owned
80.
by louiereederson 8mo ago
Wow thank you, I didn't know about this feature
81.
by louiereederson 8mo ago
Care to elaborate?
82.
by louiereederson 8mo ago
I know they acknowledge this but measuring autonomy by looking at task length of the 99.9th percentile of users is problematic. They should not be using the absolute extreme tail of usage as an indication of autonomy, it seems disingenuous.
83.
by louiereederson 8mo ago
Your ability to sign up enough customers is directly related to your ability to sustain training costs. The model runs have a short lifespan. They may serve many customers at a given point in time per run, but in order to serve those custom
84.
by louiereederson 8mo ago
Referring to my earlier comment, you need to have a model for how to account for training costs. If Anthropic stops training models now, what happens to their revenues and margins in 12 months? There's a difference between running infe
85.
by louiereederson 8mo ago
Anthropic reduced their gross margin forecast per external reporting (below) to 40%, and have exceeded internal forecasts on inference costs. This does not take into account amortized training costs which are substantial (well over 50% of r
86.
by louiereederson 8mo ago
Big tech might ahead of the rest of the economy in this experiment. Microsoft grew headcount by ~3% from June 2022 to June 2025 while revenue grew by >40%. This is admittedly weak anecdata but my subjective experience is their products s
87.
by louiereederson 8mo ago
You say this with such confidence and then ask if smaller chips require smaller wafers.
88.
by louiereederson 8mo ago
No. TSMC will not take the risk on allocating capacity to just anyone given the opportunity cost.
89.
by louiereederson 8mo ago
I think most estimates have Nvidia at more or less stable share of CoWoS capacity (around 60%), which is ~doubling in '26.
90.
by louiereederson 8mo ago
That is fair but doesn't do much to push back against the risk to the independent model vendors, particularly in consumer. This has represented a huge portion of the AI capex so far. OAI alone represented 2GW of compute in 2025. The po

More ›