$ curl shadid.dev

SHADID(1)Engineer’s Manual · Sheet 1 of 1Email Amr

Amr Shadid

Section A–A: what people use above the line, what runs beneath it below. Drag the linean A to move the cut.

I build the software people see, and everything it stands on.

By day, brain-MRI platforms and the research infrastructure under them, at NYU Abu Dhabi and Sentinel. Around that, an open-source DICOM library in pure Go, and a game that teaches the Linux kernel from the inside.

Most engineers pick a layer. This sheet shows what happens when you don’t.

2017
Building since: networks first, then software
6
Systems on this sheet, from interface to kernel
4
Platforms shipped: web, desktop, iOS, Android
2nd
Innovation Award, Abu Dhabi Global Health Week 2025 (team)

On the bench

Open items · help wanted

Selected work, in section

Fig. 1. Six systems, cut open. Pick a block to read its sheet, or run the trace.

Above grade · applications
+3 shipped appsOperations and analytics
AA
Below grade · infrastructure, protocols, kernel
$ traceroute --layers amr
  1. 1LAMINATE review screens for brain MRI
  2. 2XNAT the archive and compute behind research
  3. 3go-dicom DICOM files and DIMSE networking
  4. 4FRRouting fixes merged upstream
  5. 5Kernelfall syscalls traced into Linux
  6. 6Physical networks links, VPNs, data centers

Physical networksBedrock · Aman IT, 2017–2022

Showing LAMINATE

Sheet · LAMINATESentinelResearch, hospital pilots, and production use1 of 6

What changed in this brain since the last scan?

LAMINATE

Brain MRI analysis and longitudinal lesion assessment

LAMINATE longitudinal review: sagittal, coronal, and axial FLAIR views of a follow-up scan with lesion outlines, beside session statistics for new, enlarging, and stable lesionshospital PACSon-site DICOM gatewaylocal de-identificationprocessing servicesanalysis results returned
Fig. 2. Longitudinal lesion review, and the on-site gateway path beneath it.

Research and clinical teams need to process brain MRI, assess lesions, and compare scans across time points without stitching together separate tools.

Role. Develops LAMINATE end to end: frontend, backend, imaging integrations, and infrastructure.

  • Web and cross-platform desktop applications for medical image visualization
  • Longitudinal lesion review across time points
  • Dataset ingestion and analysis workflows with asynchronous processing
  • On-site DICOM gateways that connect hospital PACS to processing services, with local de-identification and return of results
  • Used across research, hospital pilots, and production
  • Runner-up, Innovation Award, Abu Dhabi Global Health Week 2025 (team award)

React · TypeScript · Tauri · Cornerstone.js · FastAPI · Docker

Bill of materials

Table 1. What each layer is made of, and where to see it.

LayerBuilt withSeen in
L0

Interfaces

Review screens, desktop apps, and mobile apps people work in.

  • React
  • TypeScript
  • React Native
  • Expo
  • Tauri
  • MUI
  • Bootstrap
  • Cornerstone.js
  • OHIF Viewer
L1

Services & data

APIs, asynchronous processing, and operational applications.

  • Python
  • FastAPI
  • Flask
  • Django
  • Django REST
  • Node.js
  • Express
  • Go
L2

Imaging protocols

DICOM from file to network: parsing, pixel decoding, storage, query, and retrieval.

  • DICOM
  • DIMSE
  • PACS gateways
  • De-identification
  • XNAT
L3

Imaging pipelines

Neuroimaging at scale: conversion, quality control, and preprocessing.

  • NIfTI
  • dcm2niix
  • MRIQC
  • fMRIPrep
  • QSIPrep
L4

Infrastructure

Archives, clusters, and the path from imaging data to compute.

  • Docker
  • Docker Swarm
  • Kubernetes
  • Slurm
  • HPC pipelines
  • CLIs with Ink
  • Python packages
  • npm packages
L5

Networks

Routing, links, and VPNs that keep distributed sites connected.

  • FRRouting
  • MikroTik
  • Ubiquiti
  • Zabbix
  • VPNs
  • Wireless links
L6

Kernel & systems

Linux from the shell down to the syscall, and teaching it.

  • Linux
  • C
  • Syscall tracing
  • v86

Revision history

Table 2. How a network engineer ended up in brain imaging.

RevPeriodRoleDescription
—Since inceptionTechnical LeadSentinel

Technical lead at Sentinel, the company behind LAMINATE and the upcoming CIQE platform.

  • Develops LAMINATE end to end
  • Leads engineering across LAMINATE and CIQE
C2023–nowSoftware EngineerNYU Abu Dhabi · full time

Research software and medical imaging infrastructure: the XNAT imaging archive, HPC integration, and researcher-facing tools.

  • Installs, administers, and manages XNAT, including migration from the previous system
  • Runs the local Docker Swarm deployment
  • Built custom integration with the Slurm-based HPC cluster
  • Designs CLIs and analysis pipelines for researchers
D2023–nowMobile EngineerUltimit ATS · part time

Education mobile applications for school communities, including NoorSpace, delivered on iOS and Android.

  • Shared application architecture across multiple school brands
  • Features, integrations, reliability, and releases
B2022–2023Student InternNYU Abu Dhabi · internship

Joined NYUAD’s research software work before moving into a full-time role.

A2017–2022Network EngineerAman IT · contract

Built networks and data centers and kept distributed sites connected.

  • Municipal and distributed-site networking
  • Wireless links and VPNs
  • Installing and configuring communication links
  • Network monitoring

Open source patches

Table 3. Released tools and upstream fixes.

Created · Go · open source

go-dicom

Pure Go DICOM library and CLI, no CGO. File I/O, pixel decoding, and DIMSE networking for storage, query, and retrieval. Amr’s own project, open to anyone under the MIT licence.

Also built and shipped

Practical tools for real organisations, in use.

Weight & Logistics Management

An operational tracking and reporting application covering customers, drivers, materials, vehicles, documents, and reports.

Django · TypeScript

Aman Center application

A case-data archive and analytics application helping Aman Center — The Arab Center For Safe Society organize records and understand patterns through statistical reporting.

React · TypeScript · Django

Logistics & Order Management Platform

A logistics and business-operations application covering items, orders, suppliers, vehicles, and transport pricing.

React · TypeScript · Django REST

References

Research, recognition, training, and writing.

  1. [1] Publication · co-authored abstract

    Longitudinal AI-Based MS Lesion Segmentation Tool for Heterogeneous MRI Datasets

    Multiple Sclerosis and Related Disorders, vol. 114, suppl. 1, 107459, October 2026. Contribution: platform software, internal tools, and pipeline integration.

    Read the abstract
  2. [2] Recognition · team award

    Runner-up, Innovation Award, Abu Dhabi Global Health Week 2025

    Team award for LAMINATE.

    See the announcement
  3. [3] Training

    DICOM Intensive Course, OFFIS e.V.

    Completed in June 2025.

SHADID(1)Engineer’s Manual

The whole sheet, from the command line. Type help, or pick a command below.