Skip to content
View nsssayom's full-sized avatar
☮️
Let there be code!
☮️
Let there be code!
50 followers · 76 following

Achievements

Achievement: Pull Sharkx2Achievement: Pair Extraordinairex3Achievement: YOLOAchievement: StarstruckAchievement: Arctic Code Vault Contributor

Achievements

Achievement: Pull Sharkx2Achievement: Pair Extraordinairex3Achievement: YOLOAchievement: StarstruckAchievement: Arctic Code Vault Contributor

Block or report nsssayom

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
nsssayom/README.md
Nazmus Shakib Sayom, cyber-physical systems safety and security

I’m a Computer Science PhD student at the University of Utah, in the IOTrust Lab, advised by Prof. Luis Garcia. I study cyber-physical systems safety and security, with a focus on system verification and safety falsification for robotic and autonomous systems.

Before my PhD, I developed and maintained an embedded Linux OS at meldCX. It hosted services that connected sensitive peripherals to kiosk applications across operating systems. My work included Yocto builds for enterprise ARM boards, C/C++ and Python services, security hardening, and reliable OTA updates at scale. I also taught Operating Systems and Embedded Systems.

Selected work

  • Obadh · A shared Rust engine for Bangla typing, with correction and suggestion layers, a C ABI, and WebAssembly support. Released apps for iPhone and iPad and Mac.
  • Buggy Drone · A drone simulator and reduced execution pipeline for testing a known emergency-deployment flaw, with telemetry, trace visualization, and a property-guided fuzzer.
  • World models for robotic perception · NVIDIA Cosmos video generation on HPC GPUs connected to a Webots perception-testing pipeline, with steering-error analysis and trajectory reconstruction. Course project for Prof. Guanhong Tao.
  • agent-msg · Local messaging between Claude Code and Codex sessions, with session discovery, native transports, and a SQLite journal for delivery events and correlated replies.
  • OpenGaze · A FastAPI service exposing OpenFace facial-behavior analysis, including eye-gaze and head-pose measurements.
  • DePen · A portable optical text scanner using a Raspberry Pi Zero, OpenCV, and Tesseract to recognize printed text and look up word definitions.

More projects, including Voice Remote’s transmitter and receiver, TALK-E, and Theia OMR Scanner, are on my portfolio.

Research

Property-Guided Cyber-Physical Reduction and Surrogation for Safety Analysis in Robotic Vehicles — Nazmus Shakib Sayom and Luis Garcia. SmartSP 2025, Springer LNICST 699, pp. 21–40. arXiv · Code and experiments

Honors & awards

Portfolio  ·  Google Scholar  ·  LinkedIn

Pinned Loading

  1. unmukto-org/obadh_engine unmukto-org/obadh_engine Public

    A linguistically accurate Roman to Bengali transliteration engine designed as a modern successor to Avro Phonetic Keyboard.

    Rust 8 1

  2. iotrustlab/buggy_drone iotrustlab/buggy_drone Public

    Drone simulation and reduced execution pipelines for reproducible safety analysis, with telemetry, trace visualization, and property-guided fuzzing.

    Python

  3. robotic-perception-wfm robotic-perception-wfm Public

    Robotic perception experiments connecting NVIDIA Cosmos video generation with Webots, steering-error analysis, and trajectory reconstruction.

    Python

  4. agent-msg agent-msg Public

    Native messaging between running Claude Code, Codex, and OpenCode sessions, with a shared local journal and correlated replies.

    Python

  5. OpenGaze OpenGaze Public

    Web service for facial landmark detection, head pose estimation, facial action unit recognition, and eye-gaze estimation based on OpenFace 2.0

    Python 20 2

  6. DePen DePen Public

    A Raspberry Pi Zero optical text scanner using OpenCV and Tesseract to recognize printed text and look up word definitions.

    Python 4