@hmex
Account based inPoland
About this account
- Account based in
- Poland
- Connected via
- Mexico App Store
Account-level information from X, not a live location or the device used for a specific post.
I'm an guy who has interest business, traveling, learning, meeting new people... I'm always working on something!!!
Santa Monica
Joined June 2009
- Tweets287
- Following2.6K
- Followers175
- Likes636
horacio retweeted
cool demo on how to use GEPA to take an off the shelf model like Jev (trained only on synthetic data) and tune it to your exact decision criteria simply by manipulating the prompt and labeling a few examples
you should follow Seth, he’s way ahead of the curve on AI functions
Jev is cool, but like any foundation model it needs to be calibrated to your decision criteria.
We launched jev-align: an open-source CLI to quickly teach Jev what good and bad looks like using GEPA.
Try it out! github.com/sutro-sh/jev-alig…
horacio retweeted
Google just dropped the best 1-hour course on Graph Engineering: from one agent to Loops and Graphs
00:00 – Your first AI agent
08:24 – Build agent memory
28:34 – Agentic loops
40:04 – How to build MCP
1:00:22 – Graph engineering
This 1-hour watch replaces any $500 course you could pay for.
Watch it, then try your first graph with the step-by-step guide below
horacio retweeted
Anthropic just reliesed the best 1-hour workshop on Loop engineering: from one prompt to a Loop that improves itself
5:31 - loop memory
11:02 - check, build, commit
13:06 - one feature per loop
33:03 - MVP in 1 hour
Free, the best thing on Loop engineering I've come across.
Watch it, then try your first loop with the step-by-step guide below
horacio retweeted
Google just released the best 1-hour course on Graph Engineering: from single agent to a full 24/7 system
00:00 - What Graphs are
09:16 - Build an agent
21:15 - Graph engineering explained
41:03 - Graph engineering practice
52:21 - Self improving Graphs
Free, the best thing on Graph engineering I've come across
Watch it, then build your first graph with the step-by-step guide below
horacio retweeted
🚨BREAKING: A multi-million dollar marketing agency just open-sourced its entire AI growth system.
Not prompts.
Not demos.
Real systems running real businesses.
It’s called AI Marketing Skills — built for Claude Code.
And it’s insane:
• Growth Engine → runs A/B tests using bootstrap confidence intervals + Mann-Whitney U (actual stats, not guesswork)
• Deal Resurrector → tracks leads who switch companies and re-engages them automatically
• Expert Panel → AI personas score your content until it crosses 90 quality
• ICP Learner → rewrites your ideal customer profile from real win/loss data
• RB2B Router → filters + scores + routes leads into outbound pipelines
No fluff.
No “AI hacks.”
Just… production-grade automation.
MIT licensed.
100% open source.
This is what replaces marketing teams.
Repo: github.com/ericosiu/ai-marke…
horacio retweeted
In 2007, 35-year-old Elon Musk explained how he chooses multi-billion dollar idea to work on.
He revealed:
• Why money is a bad goal
• Why competition doesn’t matter
• How great companies actually scale
12 lessons from young Musk that still feel ahead of 2026:
horacio retweeted
Game Changer: Chinese college student Guo Hangjiang (GitHub: 666ghj) codes MiroFish AI swarm engine solo in 10 days with AI assistants, explodes GitHub to 23k+ stars, bags $4.1M from Shanda Group in 24 hours, ditches dorm life to become Shanghai CEO.
Tech Highlights:
.GraphRAG builds detailed knowledge graphs (e.g., 905 entities, 3,822 relations in demos).
.Agents with Zep memory, unique traits, sims capped at ~40 rounds for efficiency.
.Hybrid Node.js/Python backend, Docker deploy, OpenAI-compatible LLMs.
.Low-resource scaling: 8B param dragon boat festival sim runs smooth (vid demo).
.AGPL license, credits CAMEL-AI, recent fixes for UI/errors.
.Online demo: 666ghj.github.io/mirofish-de… - Test predictions interactively.
Creator Quick Look:
>> BaiFu (Guo Hangjiang), BUPT senior turned Shanghai CEO at Shanda.
>> Built on 38.6k-star BettaFish & MindSpider for full data-prediction pipelines.
>> 790 contribs last year.
Real-World Plays ->
-> Sentiment: Wuhan Uni backlash sim predicts trends (repo vid).
-> Literary: Agents finish Dream of the Red Chamber's lost ending.
-> Finance: Historical data for market forecasts; like bags on Polymarket SPX bets.
-> Policy: "What-if" for bills/geopolitics.
-> Cultural: Dragon boat fest agent dynamics.
-> Ecosystems: Integrates crawlers for PR/decision tools.
Repo: github.com/666ghj/MiroFish
Mind blown: A Chinese quant college student builds an AI swarm engine in 10 days flat, explodes GitHub with 13,000+ stars, and scores $4,000,000 in funding!
Introducing MiroFish is the multi-agent simulator that's revolutionizing predictions for trading, PR, and more.
What is MiroFish?
It's a digital sandbox where thousands of AI agents with individual memories and behaviors interact like a real society.
Feed it any scenario (news leak, policy change, or even a classic novel's missing ending), and it simulates crowd reactions, debates, and outcomes to forecast real-world events.
The Creator's Story:
> In late 2025, fourth-year student Guo Hanjiang coded the core using AI assistants.
> It went viral overnight, landing him 30m Yuan (~$4m) from Shanda Group.
> He ditched the dorm, started a company, and now leads the charge.
Key Applications:
.Trading: Input financial news or reports, watch simulated market panics and price swings for predictive insights.
.PR Testing: Companies/Politics run draft statements to spot backlash and refine messaging.
.Creative Experiments: Loaded a lost-ending Chinese novel, agents role-played characters and generated a logical finale.
.Easy setup: Deploy via Docker in minutes with any LLM API key.
Pro tip: Simulate something wild like Elon Musk tweeting about Dogecoin 2.0 and spawn agent traders, influencers, and investors, generate real-time video clips of the frenzy to test moonshots or crashes risk-free.
Traders are already winning big: Check this one on Polymarket - $120,000+ net profits from spot on SPX 500 bets, powered by MiroFish sims on historical data.
His profile: polymarket.com/profile/%40mo…
For effortless gains, try Kreo copy trading: Auto-mirror pros like him and ride their edges.
Try here: kreo.app/@join
Add his wallet: [0x17559efac103ac7f361be37ec0b93888d4c55aac] to [t.me/KreoPolyBot?start=ref-j…] and start track/copy him.
Repo: github.com/666ghj/MiroFish
horacio retweeted
🚨BREAKING: Claude can now replace a $50,000 startup accelerator and advise you like a Y Combinator partner (for free).
Here are 18 Claude prompts to build and scale your tech startup:
horacio retweeted
🚨 BREAKING: The CEO of Y Combinator, Garry Tan, just open-sourced his personal AI setup.
It’s called gstack.
It turns Claude Code into a full virtual tech company:
→ CEO agent
→ Engineering Manager
→ QA tester
A conductor agent forces strategic thinking before any code is written.
Ship software like YC.
horacio retweeted
✨ You can now make TikTok carousels with Photo AI!
I like seeing @onlinedopamine's tips how to do TikTok/IG slideshows so I thought it'd be nice to add this
He says they get way more views than videos these days and if you do them well convert well also
The cool thing with Photo AI real influencers can use it to make pics with themselves in it, so it's not artificial AI characters, they can actually use it for their content
Prompts:
1) model with ponytail and white cap wearing pink string bikini, in a balinese coworking studio, bamboo walls, lush tropical plants, sunlight pouring through high windows, vibrant tropical decor, wearing relaxed business attire, thoughtful expression, soft natural lighting, candid workspace vibe. text overlay: "insane things i did as a digital nomad to become a self-employed baddie" and small letters below "(from a just graduated uni student)
2) photograph of model woman in a tranquil loft, warm amber glow from dozens of flickering candles, ornate sacred symbols etched in wood and glass, TikTok text overlay: "how i actually found myself spiritually again after being lost for years"
3) low angle view of model looking down into camera as corporate girlie in sleek high‑waist navy trousers, crisp white blouse, minimalist silver necklace, in a modern glass office. soft diffused daylight filters through windows, casting gentle shadows. background of city skyline. style: polished realism with subtle cinematic depth. TikTok text overlay: "best skill to surviving corporate?" . smaller text below (from a 5+ year corporate girlie)
This quoted post is unavailable.
horacio retweeted
i will invest $150k if you have an idea that hits 1 of these ideas.
1. you gotta build the thing
2. you will take $8k/month salary
3. i will take 33% equity
4. i will take care of the backoffice
5. i will add a $50k marketing budget
we have 6 months to hit $25k MRR
i reverse engineered 8 apps that hit $50K MRR in under 180 days and explain the pattern behind all of them
1/ find a group that spends money
2/ find a problem they repeat weekly
3/ use AI to make it work with photos/videos/inputs
4/ make the output dead-accurate
5/ replace their slow existing tools
full breakdown in the video along with my 6 mobile app frameworks and 25+ startup ideas
all free. i want to see you build some iconic apps.
i believe 2026 is an incredible time to be building mobile apps
happy building, my friends.
horacio retweeted
We replaced a $750K/year marketing team with 24 TikTok Shop AI Agents.
No actors.
No shoots.
No ghost creators.
No wasted samples.
Just 300+ shoppable TikTok videos per day going live on autopilot.
Here’s the exact workflow we use for 8-figure TikTok Shop brands 👇
Step 1 – Find what already works
Use Manus to:
- Find winning products
- Pull proven hooks & angles
Use Fastmoss + Kalodata scrape top-performing TikTok Shop videos in your niche
See which creatives actually drive GMV (not just views)
Output: A list of top hooks, angles, and formats to copy.
Step 2 – Build AI “creator” profiles
Spin up 24 AI creator accounts (we treat each as its own “agent”):
Each has a clear avatar (mom, gym bro, esthetician, student, etc.)
Each profile is positioned around one product line or problem
Output: 24 creator-style profiles ready to post shoppable content.
Step 3 – Turn winning angles into 300+ videos/day
For each Agent, we give 1 prompt that includes:
- The niche
- The brand
- The best hooks & angles from Step 1
Use AI tools to create the content:
Creatify / Arc Ads / MakeUGC → full AI UGC-style videos
Nano Banana Pro → product & lifestyle images
HeyGen is great for the advanced stuff
Output: 300+ TikTok Shop-ready videos per day
Cost per video: ~$5–$8 vs $50–$100 for human creators.
Step 4 – Autopost across a “creator swarm”
Use Reel Farm / Base44 / n8n to:
Schedule & autopost content across all 24 agents
Rotate formats (hooks, intros, CTAs) daily
Keep each account active without manual uploads
Output: Hundreds of shoppable videos posting 24/7 with no human bottleneck.
Step 5 – Double down on winners (MPS method)
Every week we:
Check which videos drove the most GMV / clicks / watch time
Clone those winners into 20–50 variations
Push them across more AI agents & more platforms (the Multi-Platform Swarm)
Everything links back to your TikTok Shop product page.
Result:
- ultra low CPMs
- 100–200 shoppable videos live per day
- GMV that compounds because posting never stops
An “always-on” creative system that doesn’t get tired or wait on creators
The $300/month stack that replaces a $50K+/month creative budget
- Manus → product research & viral hooks
- Kalodata → TikTok Shop data & creator scraping
- Fastmoss → competitor content & winner tracking
- Creatify or similar → AI UGC videos
- Nano Banana Pro → AI images
- Reel Farm / Base44 / n8n → autoposting & workflows
This is the AI Creator Agent Method we plug into 8-figure TikTok Shop brands.
I documented the full 24-Agent System (prompts, workflows, tools, dashboards).
Comment “Agent” and I’ll send you the whole thing.
(Must be connected)
PS: Repost if you want early access to the full 24-Agent TikTok Shop stack.
horacio retweeted
A must-bookmark for vibe-coders.
@YCombinator’s guide to making the most of vibe coding:
horacio retweeted
You might’ve known us as a “RAG framework” company - but we’ve been a best-in-class, agentic document OCR/workflow company for the past 1.5+ years! 📑🤖
We’re building the future of knowledge work over documents.
Our website is awesome - check it out if you haven’t already 👇
llamaindex.ai/
horacio retweeted
n8n just completely changed the game for automation builders.
You can now describe your automation in plain English and watch it build itself with their new AI Workflow Builder.
It configures the nodes and wires up the logic for you.
The learning curve just got 10x easier.
horacio retweeted
🚨 MIT just built an AI that rewrites its own code to evolve
It’s called SEAL (Self-Adapting Language Model) and it might be the first real step toward self-improving intelligence.
Instead of waiting for humans to fine-tune it, SEAL:
- reads new information
- rewrites it in its own words
- performs gradient updates *on itself*
Literally… it learns how to learn.
The results are insane:
✅ +40% boost in factual recall
✅ Outperforms GPT-4.1 using data it generated itself
✅ Masters new tasks without human input
LLMs that train themselves aren’t sci-fi anymore.
We just entered the era of self-evolving AI.
Paper: jyopari. github. io/posts/seal
horacio retweeted
If anyone needs a video guide to Karpathy's nanochat, check out Stanford's CS336!
It covers:
- Tokenization
- Resource Accounting
- Pretraining
- Finetuning (SFT/RLHF)
- Overview of Key Architectures
- Working with GPUs
- Kernels and Tritons
- Parallelism
- Scaling Laws
- Inference
- Evaluation
- Alignment
Everything you need to prepare for a job at Frontier AI Labs.
I'm taking this course and will share my learnings here on X.
Link to the playlist in the next tweet!
horacio retweeted
My WhatsApp responds to customers 24/7 now.
Built an AI assistant that handles questions, provides support, and responds in my exact brand voice.
Closed 3 deals while I was asleep last night.
No more missed messages or frantic copy-paste responses.
Customers get instant help and I wake up to satisfied clients.
Comment "CHAT" and I'll DM it to you (must be following)
horacio retweeted
3 ways to make money with AI right now:
1) AI creative agency. you build end-to-end marketing campaigns using AI. scripts, visuals, content generation, whatever they need.
2) niche tools. build AI solutions for specific markets and partner with industry experts. nobody talks about this model. dentists need different tools than lawyers.
find the niche, solve their problem, let the expert sell it.
3) AI workshops. teach companies how to use AI internally. corporate teams see AI everywhere but don't know how to apply it. charge $5K-$10K to run a workshop,
show them the tools, walk them through their workflows.
most people chase generic AI services.
the money is in going specific or teaching what you know.