Jack 🇺🇸

971 posts

Jack 🇺🇸 banner
Jack 🇺🇸

Jack 🇺🇸

@jack_dlm

AI @anduriltech, Christian, optimist

Costa Mesa, CA Katılım Kasım 2023
680 Takip Edilen286 Takipçiler
Sabitlenmiş Tweet
Jack 🇺🇸
Jack 🇺🇸@jack_dlm·
Men only want one thing and it’s disgusting
Jack 🇺🇸 tweet media
English
0
1
8
3K
Ian Laffey
Ian Laffey@ilaffey2·
the great tragedy of claude taking the fake laptop jobs from our beautiful coastal elites. this is not sarcastic btw it is really a tragedy
English
2
0
24
2.2K
Jack 🇺🇸
Jack 🇺🇸@jack_dlm·
I think AI should be a *little* sycophantic like if you are going to make humans comparatively useless at least make me feel good about it
English
0
0
0
11
Andy Fang
Andy Fang@andyfang·
Introducing Dasher Tasks Dashers can now get paid to do general tasks. We think this will be huge for building the frontier of physical intelligence. Look forward to seeing where this goes!
Andy Fang tweet media
English
303
180
3.7K
1.6M
Ian Laffey
Ian Laffey@ilaffey2·
who are the people at anthropic that worked on war claude? openai? where are the ai researchers that are pro lethality? do they exist? i’d like to meet them. pic related
Ian Laffey tweet media
English
9
2
151
9.6K
Scholar of the First Zyn
Scholar of the First Zyn@Mindrazer22·
15 dollar turkey and cheddar sandwhich in downtown Seattle bro what am I doing with my life
English
1
0
1
12
Jack 🇺🇸
Jack 🇺🇸@jack_dlm·
@thsottiaux Need a way to review changes in a parent folder of multiple git repositories, I'm rarely just working in one but still want my diffs!
English
0
0
2
310
Tibo
Tibo@thsottiaux·
What are we consistently getting wrong with codex that you wish we would improve / fix?
English
1.2K
14
873
142K
Ben Springwater
Ben Springwater@benspringwater·
Have a feeling that ‘Forward Deployed AI engineer’ is going to be one of the fastest growing new roles this year.
English
2
0
7
912
OpenAI Developers
OpenAI Developers@OpenAIDevs·
Subagents are now available in Codex. You can accelerate your workflow by spinning up specialized agents to: • Keep your main context window clean • Tackle different parts of a task in parallel • Steer individual agents as work unfolds
English
425
765
8K
1.5M
Jack 🇺🇸
Jack 🇺🇸@jack_dlm·
@Vtrivedy10 Would be most interested in continuous improvement here. Are you implying an out of the box gym environment or something else?
English
0
0
0
21
Viv
Viv@Vtrivedy10·
would love to hear what y’all are looking for and happy to prioritize! some immediate areas we’re pushing on: - async + programmatic subagents, really useful pattern to fan out massive amounts of work without blocking the main agent thread from doing work - production deployment patterns/examples, we released the langgraph cli to easy deploy but more coming on hooking up the permissions and infra (sandboxes, filesystems) when deploying - evals and continuous improvement. we’re working on polishing & publishing our suite of evals and harness eng recipes across models. basically open examples for how we measure how well models do and how we adjust the harness per model
English
1
0
3
167
Viv
Viv@Vtrivedy10·
the deepagents library is basically our starting point for doing harness engineering and shipping agents the internal agents used at the company are built on it (background coding, GTM/SDR, research) there’s primitives we find really useful across our evals and dogfooding like filesystems, multi-model, context management like compaction and large tool call offloading the goal is to give builders a good starting harness and the tools to customize and extend it to any task they want we have a lot of fun watching the open source community and customers ship with deepagents and also build evals that actually measure and improve their agents over time reach out if you’re building agents, doing harness engineering, exploring the space - we wanna help!
Hasan Toor@hasantoxr

LangChain just open-sourced a replica of Claude Code. It’s an MIT-licensed framework that recreates the core workflow behind coding agents like Claude Code but in an open system developers can inspect and modify. It is called Deep Agents. I spent a bit of time looking through the repo and it’s actually a pretty helpful reference if you’re trying to understand how these coding agents are structured. Here's what's inside: → Planning tools for breaking down tasks → File system access for reading, writing, and editing code → Shell command execution with sandboxing → Sub-agents for handling complex work in parallel → Auto-summarization when context gets too long Another useful aspect is that it’s model-agnostic, so you can plug in different LLMs and experiment with building your own coding agents on top of the same structure. If you’re exploring agent frameworks or just curious how tools like Claude Code work under the hood, this is a pretty good repo to bookmark. Link in the comments.

English
11
7
97
28.3K
Jack 🇺🇸
Jack 🇺🇸@jack_dlm·
Predicting that you will start to see some very compelling solutions for agent first integrations. If you think AI won’t automate FDEs… it will
English
0
0
0
139
urethra franklin
urethra franklin@DoodooSimpanero·
The military industrial aint really all that complex to a guy like me
English
9
266
4K
58.2K
Jack 🇺🇸
Jack 🇺🇸@jack_dlm·
idempotency is such a goated concept, gotta be one of my favorite properties of operations
English
0
0
0
30
LangChain OSS
LangChain OSS@LangChain_OSS·
Context windows are finite. Good agents know *when* to compress. We just added an autonomous context compression tool to Deep Agents (SDK + CLI) so models can trigger compaction at clean task boundaries instead of waiting for a hard token threshold. Read all about it ⬇️
Mason Daugherty@masondrxy

x.com/i/article/2031…

English
14
52
409
55.6K
Jack 🇺🇸
Jack 🇺🇸@jack_dlm·
What about recursive self-destruction
English
0
0
0
40
Jack 🇺🇸 retweetledi
Sweet Sav
Sweet Sav@lovemylife81·
American Exceptionalism - Q1, 2026
Sweet Sav tweet media
English
130
2K
27.1K
527.7K
Jack 🇺🇸
Jack 🇺🇸@jack_dlm·
@johnhawkinsrwn I think Americans having the utmost scrutiny of the employment of weapons we pay for is incredibly American and what makes us the best hegemon that has ever existed I think people care about the IRGC slaughter too but what did we expect from them?
English
0
0
0
29
John Hawkins
John Hawkins@johnhawkinsrwn·
The fact that absolutely nobody obsessed with whether the United States accidentally hit a girl's school cared at all about the Iranian regime deliberately slaughtering 30,000 of their own people a few weeks ago tells you all you need to know about their motives.
Lord Bebo@MyLordBebo

🇮🇷🇺🇸 BREAKING: Footage of the missile strike on the primary school for girls in the city of Minab. -> Cruise missile in diving mode! This proves it was the US and that the target was intentional, as the missiles goes into diving mode after finding the designated target! 168 girls aged 7-12 and teachers were killed.

English
149
133
1.6K
35.8K
Jack 🇺🇸
Jack 🇺🇸@jack_dlm·
Need a way to not get notifications for reactions in iMessage
English
0
0
1
38
Jack 🇺🇸
Jack 🇺🇸@jack_dlm·
@hpmcd1 This is what I can't seem to get a clear answer on despite all of the debate about this... it looks like it doesn't even detonate?
English
0
0
0
23
Jack 🇺🇸
Jack 🇺🇸@jack_dlm·
@RoastBeeph @Snz_BTC I agree evidence points to it being a tomahawk but have we seen any good theories for why it doesn't explode?
English
2
0
1
478
RoastBeeph
RoastBeeph@RoastBeeph·
@Snz_BTC It was a tomahawk, Christ. Been verified pretty concretely as a tomawak. This occurred before Iran was even firing back. How could it possibly be Iranian?
English
74
3
511
21.6K
Sina 🗝️⚡ BI Report
I have been silent on who actually hit the Minab school to get all the information. Now, after comparing the explosion patterns, I can confidently say this is not a Tomahawk. The evidence now points more toward an IRGC failed missile. Tomahawk carried 1000 lb (approx. 450 kg) conventional unitary warhead and doesn’t produce a small puff of dirt and dust.
Matt Tardio@angertab

The evidence is clear, this is not a Tomahawk Iran alleged that an American Tomahawk Cruise Missile hit a school (buried in an IRGC compound) in southern Iran, killing 165 people. Analysis of a newly released video tells a different story. ANALYSIS: A-I analysis confirms the wings of the munition in question sit about 40%-45% down the body of the munition. On a Tomahawk, the wings sit roughly 49%-50% down the body of the munition. The wing to body ratio of the munition in question matches an Iranian Kh-55–derived Land Attack Cruise Missile. Further, the video shows the munition in a steep dive angle for the final attack phase. This places the attack angle at approximately 70%, which is the max attack angle for a Tomahawk. The attack angle does not match the KH-55. That angle maxes out at about 55 degrees. So what would have caused this? CONCLUSION: The wing positioning alone makes the munition impossible to be a Tomahawk. The attack angle is at the max of the Tomahawk's capabilities. The typical attack angle for a Tomahawk is much lower than 70 degrees. The typical angle is between 20-45 degrees. This is due to the flight pattern of Tomahawks. They fly very low horizontally to the ground, often only 50-100 meters AGL to avoid detection and interception. In order to achieve that attack angle, the missile would have had to gain altitude several kilometers away, this would leave it vulnerable for interception. This is highly unlikely on the first day of US attacks. So what could have caused this? Simply put, GPS jamming of an Iranian KH-55. The USA and Israel were, and continue to actively jam the Iranian airspace. If the KH-55's signal was jammed, this could result in an uncontrollable dive. Think of GPS jamming more like disorienting the missile. On 03/07 President Trump stated: “No, in my opinion, based on what I’ve seen, that was done by Iran.” Today, I concur with the President.

English
701
2.4K
8.5K
1.2M