NB
Noel Biju

Noel Biju

Software Engineer
Building practical AI, robotics and developer tools.
Building Practical AI

I enjoy building software that runs locally, respects privacy, and solves practical problems.

Recently I've been exploring offline AI, robotics and machine learning.

Current Focus:Offline AI • Computer Vision • Robotics

Practical Systems Engineering & Local AI

Exploring offline AI inference, robotics sensors, PostgreSQL relational schemas, and developer tools.

Read Story & Stack

Project Evolution

A horizontal timeline mapping how each project forced me to learn deeper software engineering abstractions.

Scroll horizontally to view progression
01Project

RyMeds

Built my first large software project for pharmacy inventory & billing.

02Milestone

CRUD

Mastered transactional persistence, REST methods, and data integrity.

03Database

PostgreSQL

Deep dive into SQL schemas, ACID compliance, and concurrency locking.

04Security

Authentication

Engineered JWT authorization and database Row-Level Security (RLS).

05Project

ETLab+

Built multi-tenant institution lab manager. Won Best S3 Project.

06AI Stack

Offline AI Assistant

Discovered Ollama. Benchmarked LLaMA 3.1 & Mistral 4-bit CPU inference.

07Hardware

Robotics

IEEE internship: sensor calibration, complementary filtering, embedded C.

08Active Focus

AI Study Coach

Integrated offline LLMs with real-time learning analytics.

09Vision

Future (Edge AI)

Deploying local vision models and LLMs directly on low-power ARM robotics.

Featured Projects

Documented with empirical evidence, system architecture, and operational lessons.

View All Projects
Software Engineering2024

RyMeds

A high-reliability pharmacy management software streamlining inventory control, billing, and prescription tracking.

Read Engineering Case Study →
Full Stack2025

ETLab+

A multi-tenant college laboratory management platform with role-based access control and database isolation.

Read Engineering Case Study →
AI2025 - Present

Local AI Assistant

A privacy-first desktop AI interface running quantized open-source models completely offline.

Read Engineering Case Study →
AI2026

AI Study Coach

An intelligent study assistant transforming raw lecture notes into active recall flashcards and quiz evaluations.

Read Engineering Case Study →

GitHub Codebase & Activity

Inspect production codebases, repository metrics, commits, and language distributions.

Open GitHub Dashboard →

Latest Journal Entry

View All Journal Notes →
ArticleJuly 15, 2026

Running LLMs Locally: Lessons in Quantization and Ollama Optimization

An in-depth technical analysis of CPU-bound quantization performance, evaluating Q4_K_M vs Q8_0 GGUF inference latency on host RAM.

Direct Communication

Building an offline AI tool, robotics system, or backend service? Connect directly.