Thanh Doan

@wojakcodes what’s the current number?

if I don’t cross 5k/month passive income this year, I’m gay.

@Steve_Yegge where is antigravity / gemini cli?

I was chatting with my buddy at Google, who's been a tech director there for about 20 years, about their AI adoption. Craziest convo I've had all year. The TL;DR is that Google engineering appears to have the same AI adoption footprint as John Deere, the tractor company. Most of the industry has the same internal adoption curve: 20% agentic power users, 20% outright refusers, 60% still using Cursor or equivalent chat tool. It turns out Google has this curve too. But why is Google so... average? How is it that a handful of companies are taking off like a spaceship, and the rest, including Google, are mired in inaction? My buddy's observation was key here: There has been an industry-wide hiring freeze for 18+ months, during which time nobody has been moving jobs. So there are no clued-in people coming in from the outside to tell Google how far behind they are, how utterly mediocre they have become as an eng org. He says the problem is that they can't use Claude Code because it's the enemy, and Gemini has never been good enough to capture people's workflows like Claude has, so basically agentic coding just never really took off inside Google. They're all just plodding along, completely oblivious to what's happening out there right now. Not only is Google not able to do anything about it, they don't seem to be aware of the problem at all. I'm having major flashbacks to fifty years ago as a kid at the La Brea Tar Pits, asking, "why can't they just climb out?" My Google friend and I had this conversation over a month ago. I didn't share it because I wanted to look around a bit, and see if it's really as bad as all that. I've been talking to people from dozens of companies since then. And yeah. It's as bad as all that. Google is about average. Some companies at the bottom have near-zero AI adoption and can't even get budget for AI. They may have moats and high walls, but the horde is coming for them all the same. And then there are a few companies I've met recently who are *amazingly* leaned in to AI adoption. One category-leader company just cancelled IntelliJ for a thousand engineers. That's an incredibly bold move, one of many they're making towards agentic adoption. In my opinion, that company is setting themselves up for a _huge_ W. As for the rest, well, it's the Great Siloing. Everyone's flying blind. With nobody moving companies, no company knows where they stand on the AI adoption curve. Nobody knows how they're doing compared to everyone else. Half of them just check a box: "We enabled {Copilot/Cursor} for everyone!" Cue smug celebrations. They think this is like getting SOC2 compliance, just a thing they turn on and now it's "solved." And they don't realize that they've done effectively nothing at all. All because of a hiring freeze.

markdown is everywhere

allow grouping threads in the left sidebar by branch, under the same repo i usually work with 2 - 3 worktrees at once, and they all show up in one flat list. the annoying part is the terminal - i sometimes forget which worktree i opened a port in, so i've to check around, kill it and reopened in another one

Cursor 3 keeps getting better! What should we add next?

@trashh_dev if you open their profile you might find "prompt engineer" in the bio

i see people on linkedin posting they’re getting AI certifications. what does that even mean?

one almost full codex 5h session burns just ~10% of the weekly limit honestly codex is insanely good value tbh

@llmdevguy you can burn those tokens in 2-3 days i guess

This looks like some last-resort, budget Chinese plan. 😂 Unfortunately, they don’t have yearly plans, so you can’t take risky bet. 😂

@yacineMTB you're early

neural networks are amazing

@asaio87 it's faster for multiple experiments before ai we could only commit to one solution within a limited time, and it would eventually turn into a dead end or tech debt now i have more time to try things, compare options and iterate

I come to the conclusion that AI doesn’t really make development quicker. Only a tiny fraction of cases at most.

@jdorfman luckily it didnt ask how green in the backend was though

@capythanh This is the backend equivalent of “make the button green”

how i start most days with my agent

@JaidCodes a little bit overthinking but i rarely read generated code so it’s ok

One prompt to GPT-5.4 and the codebase is full of these for some reason.

@nkalra0123 @daniel_nguyenx try disabling the telemetry in your settings.json if you do it it the TTL will reduce from 1h -> 5min: github.com/anthropics/cla…

@daniel_nguyenx this is not true.. i just attached http toolkit on claude code.. and it shows ttl 1 hour.

It looks like Anthropic changed claude code’s cache TTL from 1h to 5m in March, causing significant quota and cost inflation.

@tomosman they're building a half of yc's startups

Anthropic are building Replit.

@emanueledpt @cjayls that’s a beauty of open source, just dropped you one star! great work

@capythanh @cjayls it’s up to you to say if it’s better than t3code i like to see it that i added more features but those wouldn’t have been possible without the strong architecture it already has

DP Code v0.0.18 just shipped! 🚀 → cleaner git action picker → better commit / push / PR flow → stronger Claude interrupt handling → cleaner interrupted states → Ctrl+C interrupt in composer on Mac → faster, stabler state syncing → less noisy tool/work logs → cleaner inline tool calls → better right-panel persistence → clearer completion/error toasts → sidebar reopen button when collapsed → project rename + Finder actions → better sidebar search → safer folder picker flow → cleaner browser panel status → better desktop browser controls → cleaner chat header actions → better picker UI → app-wide font scaling → more consistent typography → better diff code fonts → cleaner image-only messages → better queued follow-up restore → clearer git warnings/states → safer sync/pull behavior → polish across the whole app These has been a quite nice 12h of work @theo @jullerino

@cjayls @emanueledpt not better than the original, but probably better than its own previous version

@emanueledpt Do you not think calling XYZ better is subjective and also a bit disrespectful whilst tagging the guys in 😂

just dug through digital notes and found this almost a year ago: anthropic.com/engineering/cl… didnt think we'd get this far. back then `thinking` was a tool, not something native inside the models, and we were still laying the groundwork for agentic features

if you're using their models daily through any fixed subscription, the output can feel inconsistent. sometimes it's great, sometimes it isnt im on their team premium plan (= max 5x), and it's great during off-peak hours but otherwise it can get very slow / output can be worse than haiku then i switch to api usage (through openrouter) - the quality comeback, but the cost is (much) higher x.com/om_patel5/stat…

@capythanh What does that mean?

People sharing Claude Code bills with three to five zeros are scaring me. Why don't they use a monthly subscription?