Linux Dev | AI Rev Early Adopter | Gen SI\UX | Deep Agents | Droid
Phoenix, AZ
Joined July 2023
- Tweets4.5K
- Following780
- Followers2.1K
- Likes44.6K
Pinned Tweet
Bots and Dots are great for personal workflows 💯
What happens when you want to share and monetize that workflow?
That's where @LangChain Deep Agents take over - turn it into a Managed Deep Agent and scale your "dot" into a product others would love
Homie @RayFernando1337 unboxing that new MacBook
Nothing gets wasted -
I’ve requested a greasy finger on the display test - because I don’t touch monitors, but this is a classic un-boxing presentation 😂
Need It or Want It? The Touchscreen MacBook Pro xitter.twitterapi.xyz/i/broadcasts/2aWLeebPy…
3
6
594
Joshua: How are you feeling today?
Shall we play a game?
Only at @factory
Cyberattacks now outpace the defenses of even well-resourced organizations. Factory can help.
Introducing Factory Wargames: We will simulate a nation-state level attack on your company's infrastructure and cover the cost.
Apply below and if selected, we'll run the simulation, share our learnings, and Droids will help with the fix for free.
6
369
If you get to go to one conference in 2027 about Applied AI - this is the one
You’ll learn from and get to talk to @LangChain Engineers and Ambassadors who provide direct, actionable answers, not a “let’s connect”!
Join any of the included cutting edge, hands-on interactive demos and learn directly from the team who applies AI in real-business - it’s the strongest learning op in the next 6 months 🦜🔥.
You could be here!
We're bringing back Interrupt for the THIRD year
- This will be the biggest one yet (new venue that can hold ~2k people!)
- Most talks & workshops we've ever had
- New Executive Track
Register today for the early bird special: interrupt.langchain.com/
There's about 35 minutes left to register
@LangChain released their Content Studio Agent that the team uses to create the the great videos we've been seeing - Today they are presenting a webinar to cover it!
RSVP now - link below
Grab some tea, and prepare for Mind Blown 🤯
Come learn about our design agent on October 28th!
Chat with our team, see it in action, and discover how it works!
RSVP: events.langchain.com/gtm-eng…
6
384
Grok @Bot can now explain your entire ballot in plain words
Ballot Guide Bot - U.S. nonpartisan, sourced, tuned to your own values 🇺🇸
x.ai/bot/CjOZqoBE7my4zL8ql37…
What it does 👇
xitter.twitterapi.xyz/GitMaxd/status/2115477…
Less than a week to go and I started filling out my 2026 U.S. ballot and realized, I don't know anything about most of these people
40 Judges? Ballot Measures written in legalese?
Meet Ballot Guide - The Grok Bot!
Insanely smart use of @bot
“initial seed of Tasks & ideas, Grok Bot had the task of making hard, interesting, diverse evals by orchestrating a fleet of Opus & Astra agents”
Exactly what I’ve been doing to control @droid and ship using multiple computers/sandboxes -
🔖 Bot it
last week while I was fully away, experimented with pushing Grok @bot to do Eval & Environment work autonomously on my behalf.
added spend limits & gave access to infra so it could push forward without my regular intervention
- with an initial seed of Tasks & ideas, Grok Bot had the task of making hard, interesting, diverse evals by orchestrating a fleet of Opus & Astra agents
- Opus & Astra debating eval design, reward hacking, “evaluating the eval”, etc
- pulling in daily research from X into a GitHub backed brain for my review
- Giving Bot access to @modal + Codex & Claude subscriptions so it can do real rollouts on easy to use infra and fix things autonomously
won’t fully review till im back but quick glance, seeing lots of
- iterative tweaks to evals as real test rollouts come back
- self organizing behavior on fanning and debating good eval design from daily research dumps
- lots of focus on testing and trace analysis
- models can be more creative on how ti make things harder but they’re pretty good especially when forced to “just keep going”
Git Maxd retweeted
First cut: Dedicated VPS for AI.
GrokBot is using the Droid CLI perfectly well. I picked up some tips from an article by @GitMaxd. If you use Droid and GrokBot, it’s worth a read.
I’d like to get your opinion on this. Currently, my AI-related costs are:
Factory Plus ($100)
X Premium+ ($30)
Claude Pro ($20)
ChatGPT Plus ($20)
Opencode Go ($40)
VPS ($45)
I’m considering keeping Factory and canceling everything else to upgrade to X Super.
Factory already provides access to frontier models from OpenAI and Anthropic, as well as open-source (OSS) models with generous quotas, so I wouldn’t need Claude, Codex, or Opencode.
A VPS would also be entirely unnecessary, given that GrokBot runs on its own cloud-based infrastructure.
So, I’d essentially narrow it down to keeping Factory ($100) and X Super ($100). Cursor might actually meet my needs, but I find Factory’s "Harness" feature far too good to give up.
How are you all handling this? I’d love to hear your thoughts.
Polishing that @Google Map BillBoard graphic display that @FactoryAI’s @samwilhoit_ created for the epic Billboard demo 🔥
@bot driving screen caps
Droid doing @droid things
to understand the placement of the panels in 3d space for the more complex multi-panel layouts, I created a custom 3d visualizer tool to map the billboard video panels onto 3d buildings in google maps. this was a game changer compared to trying to digest 2d spec sheets for each location and adapt each panel design.
I fed the billboard spec sheets containing the exact panel sizes, placement, locations etc into @FactoryAI and enabled Google Maps 3d tiles API, and the coding agents were able to place the artwork onto the buildings in the exact configuration and adapt each creative to spec.
This is .. wtf?! 😳
@elonmusk 🏁🇺🇸
JUST IN: Elon Musk unveils the Tesla Roadster by launching it off a ramp before the car flipped in midair and safely landed on the ground.
The car is a collaboration between Tesla and SpaceX.
One of the most insane things I've ever seen.
1
10
488
Git Maxd retweeted
I've been seeing an increase in companies talk about how they have <coding agent> write all their evals automatically based on production traces
while i think this is a good start, it's very important to note that if you only ever create evals based on traces generated from your users, you'll only ever have "regression test evals".
sure, you should have some of these. but if you're doing eval dev, imo >50% of your time should be in identifying what your agent can in theory handle, and writing evals for that.
this will have overlap with what your users are asking of your agent in prod, but it'll also catch issues before your users can
you cant just turn prod traces into evals. you need a system which can also identify net-new or potentially lacking areas of your agent, and build evals there
This was an incredible demo!
The genius idea by @samwilhoit_ using Google Maps to superimpose a @FactoryAI video on a 3D real-world example Billboards rly hit 🔥
We do need to be more imaginative
Spent the day recreating a clone of what I saw - hope to share soon!
how I scaled my design workflow for a global billboard campaign that is live in NYC, Tokyo, London, and cities around the globe using coding agents in @FactoryAI 🧵🧵🧵
this thread is focused on using coding agents to reformat design assets for dozens of billboard panels that are often grouped, cropped, and placed in unexpected ways and less about designing the original motion graphics concept (which was created using a combination of generative AI video footage, and overlay graphics created with code / figma). Before coding agents, formatting all of these panels to spec would take hours / days of non-creative work...
14
639