Skip to content
View LuisEmbonStrizzi's full-sized avatar
6 followers · 4 following

Block or report LuisEmbonStrizzi

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
LuisEmbonStrizzi/README.md

Hi, I'm Luis 👋

Third-year AI Engineering student at Universidad de San Andrés (UdeSA), Buenos Aires. I build agentic systems end to end — from models implemented from scratch in NumPy to production backends powered by LLMs.

🌐 luisembon.vercel.app · ✉️ luisembonstrizzi@gmail.com

Highlights

  • 🏆 Winner — Anthropic × Kaszek × Digital House hackathon (Apr 2026): 1st place in e-commerce and 2nd overall among 55 teams. Now building Prompty, an AI listing optimizer for Mercado Libre (FastAPI · DSPy · Claude · Supabase) — in public beta.
  • 🤖 Building an agentic dispute-resolution platform for fintech: Claude agents orchestrated with n8n over Postgres/MongoDB, grounded with RAG.
  • 🚗 Multimodal trajectory prediction with a CVAE on the rounD roundabout dataset — published report (DOI) and live demo, presented at UdeSA AI Fest 2026.
  • 🩺 Applied computer vision for pediatric dermatology, in collaboration with a physician from Hospital Italiano de Buenos Aires.
  • 🧮 Machine learning from scratch — a four-repo series with every model hand-built in NumPy before reaching for a framework: housing regression, student risk classification, Kuzushiji neural nets and Fashion-MNIST unsupervised.

Portfolio source: portfolio — Next.js 16 · Tailwind v4 · shadcn/ui, deployed on Vercel.

Pinned Loading

  1. CiroVilmer/prompty-beta CiroVilmer/prompty-beta Public

    Prompt optimization and distribution platform — built with Next.js 16

    TypeScript 1

  2. cvae-trajectory-prediction cvae-trajectory-prediction Public

    Multimodal vehicle trajectory prediction in roundabouts with CVAEs — 2D interpretable latent, minADE20 0.517m, zero-shot cross-roundabout transfer (rounD). Report: doi.org/10.5281/zenodo.21365671

    Jupyter Notebook

  3. fashion-mnist-unsupervised fashion-mnist-unsupervised Public

    Unsupervised learning from scratch: PCA, k-Means, GMM and t-SNE in pure NumPy + autoencoder & β-VAE in PyTorch comparing latent spaces on Fashion-MNIST

    Jupyter Notebook

  4. housing-regression-from-scratch housing-regression-from-scratch Public

    Linear regression from scratch in NumPy on a two-market housing dataset — pseudo-inverse, gradient descent, L1/L2, stratified splits and country-segmented models (test R² 0.876)

    Jupyter Notebook

  5. kuzushiji-neural-nets kuzushiji-neural-nets Public

    MLP from scratch in NumPy (verified backprop, Adam, scheduling, batch norm) + PyTorch models with augmentation reading classical Japanese calligraphy — 0.82 test accuracy on Kuzushiji-49

    Jupyter Notebook

  6. student-risk-classification student-risk-classification Public

    Logistic regression, LDA and Random Forest from scratch — group/temporal splitting, class rebalancing and nested CV for honest student-risk metrics (test AUC-PR 0.892)

    Jupyter Notebook