When I created Claude Code as a side project back in September 2024, I had no idea it would grow to be what it is today. It is humbling to see how Claude Code has become a core dev tool for so many engineers, how enthusiastic the community is, and how people are using it for all sorts of things from coding, to devops, to research, to non-technical use cases. This technology is alien and magical, and it makes it so much easier for people to build and create. Increasingly, code is no longer the bottleneck.
A year ago, Claude struggled to generate bash commands without escaping issues. It worked for seconds or minutes at a time. We saw early signs that it may become broadly useful for coding one day.
Fast forward to today. In the last thirty days, I landed 259 PRs -- 497 commits, 40k lines added, 38k lines removed. Every single line was written by Claude Code + Opus 4.5. Claude consistently runs for minutes, hours, and days at a time (using Stop hooks). Software engineering is changing, and we are entering a new period in coding history. And we're still just getting started..
US AI companies should spend 100% of their time developing and innovating vs hanging out in DC begging for protection from competition. It’s a bad look that unveils a clear lack of self confidence.
Interesting thing: this will NOT be true for the most critical software
Like... Linux. There is no AI to write that code which has never been written before, and needs to be as efficient as possible. And several humans critiquing every single line before it's accepted
PREDICTION MARKETS OUTSHINE TRADITIONAL POLLS
Polls failed again in the US election, but Prediction markets like @Kalshi called it right from the start, showing the increasing accuracy of betting markets in election forecasts
https://t.co/BtZKUnt7bf
I've been an advocate of the betting markets.
It looks like they were the most accurate of all the election measures. They might have called all 50 states correctly (not done counting). In fairness, they did not correctly predict that Trump would win the popular vote. Nothing is perfect.
Manipulation claims were overblown, and I will still argue that manipulating betting markets has no point; it does not change the outcome and wastes money.
The Polymarket whale took home $47 million in profit. Many mocked him because he looked at the previous elections and saw polls undercounted Trump, and that was his reason for betting on him again. He was right for the right reason, as Trump was undercounted by the polls again. He told you why he was betting big, and instead of taking his lead and making money for yourself, too many laughed at him.
On top of this, criticism is being leveled against pollsters again. They have a real credibility problem.
Don't dismiss betting markets in future elections.
Think of them as I have always said; they distill all KNOWN information into one probability. They can react quickly to new news, offering a guide on how meaningful it is (like how Biden's betting was plunging DURING the June 27 debate, well ahead of his dropping out).
Betting markets are another tool for judging the outcome, a tool with a track record worthy of listening to. Do not dismiss them when you could be listening to them.
https://t.co/nvk08rcOF6
Truth is 90% of backend (web) perf engineering is adding indexes and fixing N+1s. Should be app dev 101 but it's not.
If you want to stand-out as an engineer at a fast-growing company: just find these and fix them.
From a CTO:
"I'm not sure if very junior developers should get AI tools; there's a risk that they don't learn to code, don't learn to spot the mistakes, and won't feel comfortable manually refactoring things."
My two cents: we need to stop "over-babying" less experienced devs.
@XfinitySupport why can’t I speak to an agent via the app during an outage? why do I need to turn to Twitter to get a response? Terrible customer service
BREAKING: San Francisco removes Chesa Boudin in historic recall.
See live election results, including vote breakdown by district: https://t.co/K8so2kwzTn