Skip to content
View MORSHEDMDMONOARUL's full-sized avatar
🏠
Working from home
🏠
Working from home

Block or report MORSHEDMDMONOARUL

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.

Content in all repositories owned by your account will be closed.
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
MORSHEDMDMONOARUL/README.md

MD Monoarul Morshed

AI & Edge Computing Engineer · Seoul, South Korea

Portfolio · Notes · LinkedIn · Email

Computer Science graduate, Sejong University (2026), and Teaching Assistant for Advanced Programming. I build and study systems where reliability matters as much as accuracy — my work centers on trustworthy computer vision, edge AI deployment, post-quantum cryptography, and security-first agentic systems.

Most of my projects run on real hardware rather than in notebooks: multi-camera vision on Raspberry Pi and ESP32, cryptography implemented from the standard rather than a library, and autonomous software designed around what happens when it fails.

Highlighted Projects

  • ExamShield — 16-camera AI exam proctoring system combining YOLOv8 detection with ESP32 edge hardware for real-time, on-premise monitoring.
  • PQC-Lattice-Sandbox — Zero-dependency implementation of FIPS 203 ML-KEM (Kyber), built from the standard as a study of lattice-based post-quantum cryptography.
  • Prompt-Studio — Prompt engineering workbench for building, testing, and comparing prompts against NVIDIA NIM inference endpoints.
  • PDF Quiz Generator — Flask and LLM application that turns any document into a summary, a generated quiz, and a chat interface.

Focus

Area What I work on
Trustworthy computer vision Detection for monitored environments, interpretable alerts, false-positive control
Local & edge AI Inference on constrained devices, Raspberry Pi and ESP32 deployment, NPU-oriented optimization
Post-quantum cryptography Lattice-based key encapsulation, implemented and validated from the specification
Security-first agentic systems Capability gating, taint tracking, journaled recovery, human-in-the-loop control

Core Stack

Python PyTorch OpenCV React Rust

Elsewhere

Portfolio and engineering notes: mdmonoarulmorshed.com · Languages: English (fluent), Bengali (native), Korean (conversational)

Pinned Loading

  1. ExamShield ExamShield Public

    AI-powered exam proctoring: YOLOv8 computer vision with ESP32 IoT sensing, scaling to 16 cameras on Raspberry Pi edge nodes.

    Python 2

  2. PDF_Quiz_Generator PDF_Quiz_Generator Public

    Flask and LLM web app that turns any PDF into a summary, an auto-generated multiple-choice quiz, and a chat-with-your-document interface.

    Python 1

  3. PQC-Lattice-Sandbox-ML-KEM-Kyber-Architecture- PQC-Lattice-Sandbox-ML-KEM-Kyber-Architecture- Public

    Zero-dependency Python implementation of NIST ML-KEM (Kyber) across all FIPS 203 parameter sets, with a custom NTT and 25 deterministic validation tests.

    Python 1

  4. Prompt-Studio Prompt-Studio Public

    Prompt engineering workbench that turns rough ideas into structured, token-optimized specifications using NVIDIA NIM, with quality scoring and a local fallback optimizer.

    Python 1