Slashy reminded @mynameisyahia twice to follow up with someone he'd overlooked in his inbox. It was a large company. Now they’re a customer of @getcontextdev
Before Slashy he used to do everything in Gmail. Until he realized he was waking up to 400 emails, and taking 3 hours just to get through 100 of them.
He tried it skeptically. Next day he cleared 500 emails in an hour, every one a personal reply that sounded like him, because Slashy had learned from what he'd already sent.
Now he just spends 30 minutes a day on email. His only concern?
"It might be a little too magical. I don't know what's going on, but it works."
Fun fact: @slashyemail was my first YC-backed customer, 6 months before I even applied to YC.
Then I got into YC, my inbox exploded, and my response SLA started slipping
So I called Harsha (happens to be my neighbor now) and asked him to come by and help me set up Slashy
As a product-led company, I get A LOT of email every day. I try to stay on top of it while still keeping every reply personal
And when I send an email that needs a response, it reminds me to follow up a few days later (GOATED feature)
I really believe in this company, cool founders, and they let me terrorize them over iMessage whenever I want a new feature.
I talked to a founder today who's doing a software startup after two previous hardware startups. He's so much more relaxed that he seems a different person. Hardware is hard. It's at least possible now, but it's still hard.
The startups YC funds now are much more serious than they were in the supposedly good old days. In the last couple days I've met with startups making optical switches, rewriting the software infrastructure of manufacturing, making nuclear reactors, and curing cancer.
@ArthurMacwaters@AlesyaMacWaters Not waiting in the house and instead waiting in the car is the play.
I go to the car when waiting for @nathaliahindle and just start catching up on my emails while I wait.
Hermes Agent Tip of the Day: 🪽
If you’re running GPT-5.6 with Hermes, there’s a brand new setting you should know about.
Hermes can now use OpenAI’s native Responses API server-side compaction for long-running sessions.
Turn it on:
compression:
codex_responses_native: true
Here’s the cool part:
Instead of waiting for Hermes to locally compress a massive conversation, OpenAI can compact the older context server-side first.
Hermes then keeps the encrypted compaction checkpoint and replays it on future turns.
So you get:
→ Better long-session context management
→ Persistent compaction across turns
→ Less unnecessary transcript baggage
→ Hermes’ normal local compression still sitting there as a fallback
And this isn’t blindly enabled across every provider.
It’s specifically gated to GPT-5.6 running through the direct OpenAI API or your ChatGPT Codex subscription.
OpenRouter, xAI, Copilot, local models, etc. don’t get the field shoved at them and told “good luck.” 😂
This one landed yesterday.
If you’re using Hermes for long-running coding or agent workflows with GPT-5.6, I’d absolutely be testing this. 🤘🏻
We’re still a small company, serving 500–1K API requests/min basically 24/7
There is a 0% chance I’m yeeting code into prod without reading it first
The number of silly mistakes I’ve seen SOTA models make just in the last week has me feeling quite uneasy
This week’s Paper Club is all about robotics. Every year for the last decade, someone has promised that the era of robotics is just around the corner. But we’re still waiting.
So we gathered a bunch of the top researchers working in AI and robotics to present the latest findings on where we are and what comes next.
Thanks to the following presenters:
0:00 – @FrancoisChauba1: Ten years of “next year, robotics is solved”
7:59 – @marceltornev: MEM - Multi-Scale Embodied Memory for Vision Language Action Models (https://t.co/czkSDRpaGc)
20:21 – Milan Ganai: Self-Supervised Bootstrapping of Action-Predictive Embodied Reasoning (https://t.co/YovZkHCoB7)
33:42 – @tylerlum23: SimToolReal - An Object-Centric Policy for Zero-Shot Dexterous Tool Manipulation (https://t.co/a44yRaeu14)
51:21 – @NikolausWest: Why the next great robotics companies will start with teleoperation
1:08:30 – @BillJiao930 & @Guanming717: World action models and what comes after VLAs
I asked a startup today how fast they were growing. They said the current number was meaningless because it was just a spike from launching something new. So I asked how fast they'd been growing before the spike. 40% per week!
Peter Thiel described this exact thing in 2020 email to the Meta board:
If housing is too unaffordable, young people will be excluded from the capitalist system and will eventually turn against it
This is happening in real time in NYC and California
If SF ever becomes 'cool' in the traditional sense it will signify the full dilution of the 'weirdos' that made (and make) it awesome & successful.