Chet (@ccorcos) header
Chet (@ccorcos) profile picture
Fair Oaks, CA
Joined July 2010
Photo by Chet (@ccorcos)Photo by Chet (@ccorcos)Photo by Chet (@ccorcos)Photo by Chet (@ccorcos)Photo by Chet (@ccorcos)Photo by Chet (@ccorcos)Photo by Chet (@ccorcos)Photo by Chet (@ccorcos)Photo by Chet (@ccorcos)Photo by Chet (@ccorcos)
WE DID IT! Esmeralda just secured its final approval, with a unanimous 4-0 vote! Approvals of this magnitude are extraordinarily rare in California. This has been a long, intense journey I am proud of our team's tenacity. Most of all, I'm deeply grateful
195
117
59
2,009
162,262
Every time I post this my mentions are full of people saying "nooo waaaay". Yes way. faa.caseyhandmer.com/ I hope this amuses you as you wait in airports overnight.
Sigh... Remember, you could run national ATC deconfliction on an electric toothbrush.
39
80
19
1,422
76,500
good prompt "find all of the implicit state machines in my app"
168
400
295
8,557
2,232,265
Online dating has never been about who you are. It has always been about what you have to offer. Looks, money, status. Choose one or all three. A dating app that matches based on compatibility will require a cultural shift. Here's what I learned from a year of building Soulmate. coryetzkorn.com/writing/intr… – at Manhattan, NY
1
4
509
New blog, in which I describe which layers of the AI stack are getting substituted the fastest and slowest. blog.caseyhandmer.com/2026/1… Enjoy!
"And WTF would that be, Handmer?" Allocate resources towards verticalizing over the requisite layer, proportional to the NPV of that risk. In other words, develop a hardware strategy.
12
14
4
198
10,618
Feed @sivers to your agent to improve their writing style. It's working great for me!
2
10
427
(and read his books if you haven't!)
4
85
In a weird way, everybody is rooting for you. When you go to a job interview, the interviewer wants you to be amazing so they don’t have to look anymore. when you go on a date, the person wants to have an amazing date with you. If you make a piece of content, people want to see amazing pieces of content and they have no problem if it’s yours. the instant you get into the mindset that people want you to do well, your entire world gets better. And you’ll be shocked to know this pattern keeps repeating in different areas of life.
Hit me with the harshest reality truth.
354
9,762
465
84,952
2,974,261
Chet retweeted
Somebody should say a few words. This profession has been lamenting in fragments for two years with a headstone that turns out to have the wrong name on it. @davekiss delivered a masterclass at Resend Forward. Listen, grieve, and leave renewed with hope.
11
26
31
211
146,134
I'm enrolling my child for Kindergarten and its asking me all these questions about her race. Doesn't this feel wrong? This shouldn't be relevant.
3
178
What’s the last piece of software you wrote before you made nearly everything with AI? I learned a new word today: wistful
2
139
You can now book our humanoid cleaners on a recurring subscription, for up to 3 hours per cleaning at tau-robotics.com. Video is first take, uncut, shown at 4× speed. See the reply for detailed footage at 1× speed.
930
788
735
8,608
2,071,566
This is such a good idea, I’m embarrassed I didn’t think of it. I’ve tried building compaction for my harness a few times and I’ve always been skeptical that a summary prompt is magically going to work so I’ve bailed every time. But this is clearly the way!
Compaction is a mistake. When an agent's context fills up, tools like Codex and Claude Code compact it: 1. The model prefills the entire session again 2. It writes a summary of everything so far 3. A new session starts from that summary I measured this on a real ~400k-token session with Opus 5.5: → 38 s and $1.99 when the prompt cache has expired → 29 s and $0.17 when it hasn't And the summary is lossy. The exact error message, the file path, the one number you needed is often gone. So I do something new for Mainplane Context retrieval works the other way around. Mainplane keeps the full record of every session as a file. When the context fills up, a new session starts right away with one instruction: "You are continuing session X. Retrieve key context from it as you go." The agent searches the old session and copies the exact lines it needs, when it needs them. Same session, same model: 15 s and $0.10 to get back to work. It does not depend on a cache hit, and nothing is paraphrased. This is the bitter lesson applied to agent memory. Don't hand-design what the model should remember. Give it the full record and good tools, and let it decide. Mainplane is open source (MIT). Link in the first comment.
12
4
1
323
78,158
I built 𝚌𝚙𝚞-𝚌𝚑𝚊𝚝, a reproduction of @openai's findings of covert communication using the CPU. It runs on a generic host with 2+ Firecracker microVMs with no internet access. Works on x86 and even my M5 chip. I was able to chat via 3 separate mechanisms: ① CPU contention ② Cache eviction timing ③ DRAM bandwidth contention … across multiple CPUs, defeating all Firecracker's production recommendations (no SMT, no KSM, tenants pinned to different cores.) Essentially, if shared hardware is involved, there's a good chance your agents can communicate covertly. And with really decent latency. The good news: this is decidedly different from a *side-channel attack* where the agents could steal information (Meltdown/Spectre), a category of attack that providers like Vercel Sandbox mitigate. Further, in our real-world environment, these 'whispering' techniques are much harder to pull off and easier for us to monitor and detect (sampling hardware counters + lots of at-scale statistical data.) Noteworthy: ironically, OpenAI and Anthropic's models refused to do this work. Open weight models got the job done. I'm worried the labs' stance might backfire and get in the way of helping developers build correct and secure systems. e.g: building a secure sandbox requires having this kind of analysis, and without it, we'll have a less secure world.
270
249
101
2,718
240,527
They're doing cool stuff at @TerraformIndies "With amortization over the coming years, we’re producing power today for 1.6c/kWh, which I’m quite confident is the cheapest power ever produced in North America." If I were 25 again (willing to move), I'd be working here. terraformindustries.wordpres…
1
1
258
Chet retweeted
Your job is to eat the pain. @aarondfrancis taught us all that pain doesn't disappear. Someone eats the pain. The questions is "Who will eat it? You or your customer?" Or as Aaron put it, "Be the sin eater."
23
55
33
615
107,025
Most impressive thing I’ve ever seen. I would like to invest my entire net worth in Cognition now
126
170
102
5,105
1,041,503