Shaik Kareem
AI/ML Developer • Full-Stack Developer
About Me
AI/ML‑focused developer who loves building intelligent systems. I design and integrate models with scalable back ends and modern UIs — with strong DSA in Java.
AI/ML First
I build intelligent features (Prompting, LLMs, RAG, analytics) and integrate them with robust back ends.
Product-Focused
I enjoy shipping full-stack apps end-to-end with clean APIs, auth, and DX-friendly tooling.
Performance & Realtime
Care about responsive UIs, data fetching patterns, and realtime collaboration where it matters.
What I’m exploring
Model serving, vector search, and production-grade data pipelines.
Technical Skills
Programming Languages
Frontend Development
Backend Development
Databases & ORMs
AI/ML
Business & Analytics
Developer Tools
Core Computer Science
Projects
Foreknow – AI-Powered Cognitive Assessment Platform
Built a full-stack platform with interactive memory, Stroop, image recall, and speech-based tests for early dementia detection. Integrated a multi-agent AI pipeline (CrewAI) orchestrating Gemini 2.0 Flash, Whisper, and Hugging Face models to generate diagnostic reports, summaries, and email communication. Automated test metric analysis, reporting workflows, and responsive PDF exports within a polished Next.js experience.
Hand Gestures Controller
A system for hands-free computer control using OpenCV for video capture and Mediapipe for high-fidelity hand tracking. Translates gestures into real-time keyboard and mouse commands.
Aizy - Full Stack AI Platform
An AI-driven e-commerce platform with a Next.js frontend, Express.js backend, and a Python/FastAPI microservice for ML-powered fashion recommendations. Features a hybrid MongoDB and MySQL database architecture.
Collaborative Text Editor
A real-time, multi-user text editor built from scratch using Java Sockets and NIO. Implemented a custom TCP protocol for live text synchronization and typing indicators, with a Java Swing GUI.
SehatSaathi 2.0 — AI Health Bot
An Azure Health Bot–powered medical assistant with a modern Next.js UI. Implements conversational symptom triage, intents, protected routes, and responsive design; deployed for demo and shortlisted in SIH.
3D Portfolio Generator
Interactive 3D portfolio experience built with Next.js, Three.js, and React Three Fiber. Integrates Google OAuth and GitHub APIs to generate personalized, animated portfolio scenes. Built for #GitHubCopilotChallenge.
Space Debris Tracker
Orbital-tracking utility that consumes TLE data and computes satellite/space-debris positions with Skyfield. Provides time-based propagation and basic visualization utilities for analysis.
Deepseek — VS Code Extension
A developer productivity extension integrating DeepSeek capabilities into VS Code. Demonstrates extension activation, command palette integration, and API-driven features.
Achievements
My Journey
Born
Born on 12-07-2006.