Skip to content
View yves-a's full-sized avatar

Highlights

  • Pro

Block or report yves-a

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 supported. This note will be visible to only you.
Report abuse

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

Report abuse
yves-a/README.md

Yves Alikalfic

I like to build products — full-stack • AI-adjacent • practical UX
Resume LinkedIn Email


About

CS-minded builder focused on shipping small, useful products. Recently:

  • Autonomous Agents: Built an NBA Front Office agent that evaluates trades and roster construction.
  • ML Fundamentals: Implementing core algorithms from scratch, and watching Andrej Karpathy's Neural Networks course
  • System Architecture: Designing multi-service applications using Java and Python with Docker orchestration.

I value clean READMEs, reproducible setups, and small demos people can actually run.


Tech stack

Languages: TypeScript/JavaScript, Python, Java, Go, Swift
Frontend: React, Vite, HTML/CSS
Backend: FastAPI, Flask, Node, Java (Maven)
Infra/Dev: Docker & Compose, Nginx, GitHub Actions
Data/AI: LLM Orchestration, Embeddings (Ollama/OpenAI), Scikit-learn


Highlights

1) LeGM — NBA Front Office Agent (Python, OpenAI, NBA API)

Repo: LeGM
What it does: An autonomous AI agent that acts as an NBA General Manager. It analyzes real-time player data, evaluates trade fairness using custom logic, and provides strategic roster recommendations.
Key Features: Multi-step reasoning for trade analysis, real-time data fetching via NBA API, and "Phase 3" scouting logic.
Stack: Python, OpenAI SDK, Pandas, NBA_API.


2) Simple ATS — Java + Python + React, Dockerized

Repo: simple-ats
What it does: Compares resumes and job descriptions via embeddings with a Java service calling a Python FastAPI service; dev/prod via Docker Compose.
Stack: Java (Maven), Python (FastAPI), TypeScript, Docker, Nginx, Ollama.
Notes: Demonstrates cross-language service communication and container orchestration.


3) Sift - Gift Recommender — web + mobile (K-means, RN swipe UI)

Repo: sift
What it does: Personalized gift suggestions via K-means on user preferences; shipped as web + React Native app. Drove 60 sales from 1.3k+ users and lifted engagement 1.75× after adding a Tinder-style swipe UI.
Good for: Recommender systems, RN gesture UX, and real-world user growth.


4) FantasyAI — Predictive Analytics for Fantasy Sports

Repo: fantasyAI
What it does: Leveraging machine learning to predict player performance and optimize fantasy lineups.
Focus: Data engineering pipelines, regression models, and sports analytics.
Stack: Python, Scikit-learn.


Quick links


Open to

New Grad Software Engineering roles (Full-stack/Backend/AI), starting May 2026.
For a quick repo tour or feedback, email [email protected].

Pinned Loading

  1. simple-ats simple-ats Public

    A simple Applicant Tracking System (ATS) that compares resumes to job descriptions using AI embeddings.

    TypeScript

  2. sift sift Public

    Swipe-to-Discover Gift Recommender

    JavaScript

  3. fantasyAI fantasyAI Public

    Python

  4. LeGM LeGM Public

    building an nba gm agent

    Python