Aaron 🛠️🤖🌎

Own your mind. Own your work. Own your stack. Ex-agency owner, 20+ yrs in tech. Building apps, local AI and a side of travel. Exit the default.

[Aaron 🛠️🤖🌎 @localhostnomad]

This is what going solo at 40 feels like.

I’m Aaron. 20+ years in tech. Ran a marketing agency for 8 years. Built apps and websites as a freelancer along with almost all of our agency systems.

In August, my business partner left our SaaS startup.

Now it’s just me, rebuilding on X while traveling across South and Central America.

What I’ll share:

→ build lean with AI
→ run AI on your own hardware
→ live and work from anywhere

What are you building, or trying to get built?

Lets connect and grow together!

[Aaron 🛠️🤖🌎 @localhostnomad]

@cesarsuarezpab Incredible!

[Aaron 🛠️🤖🌎 @localhostnomad]

@DadOmega_ @analogalok They get paid from when they host the model on their subscriptions and use that data to improve the open model usually

[Aaron 🛠️🤖🌎 @localhostnomad]

@ajaxships Bad feedback probably. I love to design things for me

[Aaron 🛠️🤖🌎 @localhostnomad]

@coffeecup2020 Where you testing it?

[Aaron 🛠️🤖🌎 @localhostnomad]

“Aren’t you scared to go to South or Central America?”

The answer is a resounding no:

- It’s no more scary than much of the US
- There’s no lack of McDonalds 😂

Of course it depends on what part you visit, but Panama is super safe and a fairly wealthy country with the canal.

Cost of living for most things is still less than the US, especially Uber rides and services so your US dollar goes further.

While it’s not as cheap as Colombia/Ecuador - it’s a great entry for any expat trying to adjust to life outside of the US with still maintaining a close proximity.

Where’s your favorite place to travel or live?

[Aaron 🛠️🤖🌎 @localhostnomad]

@Rubio_tv Nice to meet you Rubio! Just shot you a follow.

[Aaron 🛠️🤖🌎 @localhostnomad]

This is what going solo at 40 feels like.

I’m Aaron. 20+ years in tech. Ran a marketing agency for 8 years. Built apps and websites as a freelancer along with almost all of our agency systems.

In August, my business partner left our SaaS startup.

Now it’s just me, rebuilding on X while traveling across South and Central America.

What I’ll share:

→ build lean with AI
→ run AI on your own hardware
→ live and work from anywhere

What are you building, or trying to get built?

Lets connect and grow together!

[Aaron 🛠️🤖🌎 @localhostnomad]

@josua_zentgraf 100% - people will spend all their time taking any the what if’s and then they never actually capitalize on the present moment and the opportunities that exist now

[Aaron 🛠️🤖🌎 @localhostnomad]

If you want to follow my AI model training go here: https://t.co/82RXgXOTWx

[Aaron 🛠️🤖🌎 @localhostnomad]

I have to agree with Andrej…

I trained my 2B model for just for 3 days and many of its specialized security code review tasks are already running better than Deepseek 4.1 Flash 1017

Smaller models makes much more sense from a standard hardware perspective bringing processing to edge devices.

I am lucky to have 2 DGX sparks with 256gb unified memory and a MacBook Pro with 48gb but the majority is working below 32gb of memory or vram.

A lot of the frontier models are a waste and overkill for many specialized tasks that businesses need.

The future is local AI

[Codez @0xCodez]

Andrej Karpathy:

“A 1B model can be much more effective than most current LLMs.

90% of current models waste a ton of capacity and money on remembering stuff that doesn’t matter.”

here is his 5 core thesis about Small LLM’s:

1. the intelligence was never the knowledge. Strip the memorized internet, keep the reasoning. What’s left is the cognitive core, and it’s small.

2. the core might fit in ~1B params. Maybe less. “Plenty of room at the bottom.” We’ve been paying for a library when we needed a brain.

3. size stopped being the bottleneck. What makes an agent work is the tools it can call and how tight the feedback loop is, not parameter count.

4. intelligence lives in the system, not the weights. A raw LLM is a CPU with no OS. Small core + good harness > giant model alone.

5. you can own it. A model on your hardware is one no one can rate-limit, price-hike, or deprecate on you.

in 12 months the gap between teams shipping on-device and teams paying per-token frontier bills won’t be capability.

TwIL an 8.8B model at 90%+ of frontier isn’t a benchmark flex.

It’s the thesis Karpathy’s been repeating all year, made real.

study more on this below ↓

[Aaron 🛠️🤖🌎 @localhostnomad]

@robj3d3 Should have just asked REA to reverse engineer your AirPods problem and fix them instead 😂

[Aaron 🛠️🤖🌎 @localhostnomad]

Qwen 4 is coming.
Qwen 4 is coming.
Qwen 4 is coming.
Qwen 4 is coming.

[Qwen @Alibaba_Qwen]

The becoming begins.

[Aaron 🛠️🤖🌎 @localhostnomad]

@Reelcinestudio Voice over, music team were generated separately but ltx had the SFX. Claude edited it together.

[Aaron 🛠️🤖🌎 @localhostnomad]

4-minute AI short film. Directed by Claude, rendered on my desk, $0 in API fees.

FLUX 3 Klein → frames
LTX-2.5 → video + SFX
Fish Audio Drama 3 → voice
ACE-Step → music

2x DGX Spark + a MacBook

[Aaron 🛠️🤖🌎 @localhostnomad]

@OrangeBlimpTV Yeah its wild to think we have the ability to do all this now locally

[Aaron 🛠️🤖🌎 @localhostnomad]

@NickyLi42 It’s pretty cool, the models keep getting better and better

[Aaron 🛠️🤖🌎 @localhostnomad]

@UrbanAstroFella Haha

[Aaron 🛠️🤖🌎 @localhostnomad]

Distilling DeepSeek V4.1 Flash 1017 into a 2B security model on 2 DGX Sparks, one night at a time.

Gap to the teacher: 11 → 6 points in one night
Real CVE bugs caught: 2 → 9
67 ms per answer vs 25 s

Retrains nightly at 3:45 AM EST. Follow the race 🚂

[Aaron 🛠️🤖🌎 @localhostnomad]

@master_jpma Done! Looking forward to following your journey

[Aaron 🛠️🤖🌎 @localhostnomad]

@yushbuilds Likewise!

[Aaron 🛠️🤖🌎 @localhostnomad]

@EmiliOlivi_ai Thank you!