Christian Eilers (@TatanDev) profile picture

Ingeniero de Software. @X @SpaceXAI

Joined August 2023
Photo by Christian Eilers (@TatanDev)Photo by Christian Eilers (@TatanDev)Photo by Christian Eilers (@TatanDev)Photo by Christian Eilers (@TatanDev)Photo by Christian Eilers (@TatanDev)Photo by Christian Eilers (@TatanDev)Photo by Christian Eilers (@TatanDev)Photo by Christian Eilers (@TatanDev)Photo by Christian Eilers (@TatanDev)Photo by Christian Eilers (@TatanDev)
Use XChat
A student-organized tech conference from my home country sent me a terribly long email inviting me to be a speaker at their 2027 edition. I discarded the email, it seemed like a bad event based on the email. Later, that same student DM'd me on X asking me if had seen the email and whether I was interested in the event. I replied and explained to him why I had disregarded the email. We talked a little bit over X DM's about the event, and how I could be helpful. I ended up accepting the speaker slot! 🚀 1. Don't write shitty cold emails (this wasn't even AI slop, it was just terse). 2. Just DM people on X!
5
360
The user seems frustrated
1
62
I agree to an extent, I think when needing to be smart about developer time and token usage its better to combine a bunch of PRs together to run one round against all of them together
maybe a hot take but i now consider adversarial reviews mostly a waste of tokens. far better to use those tokens on empirical combined with formal verification. i have updated pstack to use the interrogate skill less, and eventually i think i may remove it altogether i have found very little value with code review in the abstract, agents tend to just get lost in the nitpicks and yagnis
2
5
696
Then again, my opinions on this are changing day by day from experimentation, so its hard to think theres a "best approach" to things. But verification is for sure an important factor
1
33
My brain is in a constant fight of wanting to do more cause its exciting and wanting to rest cause its tired
12
355
We need more massive and cheaper inference
2
139
it all comes down to sensors and actions. With proper data on what is occurring and proper ways to act upon that data, everything in between is just an optimization problem
4
302
agents should deploy to prod
2
1
6
442
Christian Eilers retweeted
Hope you'll find this useful, on the only platform that does not charge a revenue sharing fee for paywalled content 💪. Nice work @saurinx @YanhuaLiu2n @allegrajacchia on top of the XChat foundations laid by @cambridgemike @skunichi @antonlikespizza @TatanDev @twericz.
Introducing Subscriber-only Group Chats. Chat in real time with your most loyal supporters and build a community that keeps them subscribed. Now in beta for creators with 800 or fewer subscribers. Set yours up in Creator Studio > Subscriptions.
15
6
93
4,487
The goal is to automate everything
1
181
Christian Eilers retweeted
I see something on X, I click share -> Send to chat -> Select Grok Bot -> buy this. Done!
20
4
4
215
22,019
getting gold is not even the exciting part here. It's the potential to generate rare metals that are useful in engineering at a low cost
Last month, we demonstrated high-rate alchemy: turning mercury into real, stable gold, beating CERN’s record by a factor of 10. Last week, we enriched isotopes in the world’s most powerful plasma centrifuge, beating the Soviet record that stood for 40 years. Now, we’re building a fusion reactor in San Francisco to achieve record power density.
62
"Oh you use AI so you dont have to think?" No bro I use it to parallelize 20x and be 10x faster at building stuff
2
1
7
195
Math proofs have been aided by computers for decades now and no one has complained mathematicians don't have to think because of it
2
25
From the past months of experimenting with being fast with agents I've learned the following: - Your agent must be able to navigate and control the app the same way any human would - Your agent must be able to debug everything the way a human would. This means every feature must have proper metrics, proper tracing and proper logging. - Once you start parallelizing enough you ideally want a global overview of all the features that its building in parallel and what stage they are in (implement, review, babysit) - The best way of getting review is with a bot that will isolate an environment to test everything in
2
116
Christian Eilers retweeted
okay the deskbot is looking cute 😵‍💫 @grok @bot
26
44
10
710
20,363
build build build. ship ship ship.
1
1
172