Abdelrhman Mohamed.
11 open-source projects, 15 merged pull requests
Software engineer

Abdelrhman Mohamed

Software Engineer focused on designing and building systems that keep working as load grows, with a strong emphasis on correctness, maintainability and real-world engineering constraints.

Interested in software architecture, distributed systems, performance, security, and the engineering practices that turn complex requirements into tested, deployable software.

View My Work
Oversold seats, 100-VU inventory burst
0
Eventora · local, Docker Compose; 8-seat tier
Requests in five capacity runs, 0 failed
569,066
Eventora · local, Docker Compose; 0 server errors
Read ceiling, one 1-CPU replica
660 req/s
Eventora · local, Docker Compose; read path, 0 errors
Automated tests
228
Eventora · 228 / 228 passing

01 / About

About

Stack, by where it was used
Backend
  • Java 21Used in: Eventora
  • Spring Boot 3.5Used in: Eventora
  • Spring Security (JWT/RBAC)Used in: Eventora
  • Spring Data JPAUsed in: Eventora
  • Spring State MachineUsed in: Eventora
  • FlywayUsed in: Eventora
  • PythonUsed in: Recruiter-Pro, SysPlex, LexIntelligence, PhishSniffer
  • FastAPIUsed in: Recruiter-Pro, SysPlex, LexIntelligence
  • FlaskUsed in: SysPlex
  • GoUsed in: SysPlex
Data and messaging
  • PostgreSQLUsed in: Eventora
  • RedisUsed in: Eventora
  • RabbitMQUsed in: Eventora
Delivery and testing
  • DockerUsed in: Eventora, SysPlex
  • Docker ComposeUsed in: Eventora
  • NginxUsed in: Eventora
  • GitHub ActionsUsed in: Eventora, Recruiter-Pro
  • k6Used in: Eventora
  • TestcontainersUsed in: Eventora
  • JaCoCoUsed in: Eventora
  • pytestUsed in: Recruiter-Pro, SysPlex
AI services
  • scikit-learnUsed in: Recruiter-Pro, PhishSniffer
  • Ollama (HTTP)Used in: Recruiter-Pro
  • ChromaDB + LangChainUsed in: LexIntelligence
  • PyTorch· notebookUsed in: PlantVillage notebook
  • StreamlitUsed in: PhishSniffer
Frontend
  • Next.jsUsed in: Eventora, Recruiter-Pro
  • ReactUsed in: Eventora, Recruiter-Pro
  • TypeScriptUsed in: Eventora, Recruiter-Pro
  • Tailwind CSSUsed in: Eventora
  • VitestUsed in: Eventora

02 / Work

Work

Event search results filtered by text, city, category and date, light theme

Fig. 01 / 06 · Event search results filtered by text, city, category and date, light theme

Flagship project

Eventora

Event ticketing platform: one Spring Boot deployable (modular monolith) and a Next.js frontend.

  • Java 21
  • Spring Boot 3.5.13
  • Spring State Machine 4.0.0
  • PostgreSQL 17
  • Redis 7
Oversold seats, 100-VU inventory burstlocal, Docker Compose; 8-seat tier
0
Requests in five capacity runs, 0 failedlocal, Docker Compose; 0 server errors
569,066
Read ceiling, one 1-CPU replicalocal, Docker Compose; read path, 0 errors
660 req/s

Case study: EventoraSource: Eventora (opens an external site)

Recruiter-Pro dashboard with the upload area and the four-step processing pipeline

Fig. 01 / 04 · Recruiter-Pro dashboard with the upload area and the four-step processing pipeline

Case study

Recruiter-Pro

CV screening against a job corpus: parse, extract, score, explain.

  • Python
  • FastAPI 0.111.0
  • scikit-learn
  • Ollama (HTTP)
  • pytest
Jobs in corpus
800
Branch coverageCI floor 81%
84.07 %
TestsREADME states 544 and 530 in different places
500+

Case study: Recruiter-ProSource: Recruiter-Pro (opens an external site)

Case study

SysPlex

Cross-platform system observability: a Go native agent, a Bash agent with a FastAPI API, a PowerShell collector for Windows, and a Flask dashboard that polls every 2 s.

  • Go 1.21
  • gopsutil 3.23.11
  • Python 3.11
  • FastAPI
  • Flask 3
Dashboard refresh
2 s
Docker Hub pulls
124
pytest tests
104

Case study: SysPlexSource: SysPlex (opens an external site)

Also built

  • LexIntelligence

    Dual-model RAG assistant for legal documents.

    Scenarios completedself-run evaluation
    10/10
    Faithfulnessself-run, small manual sample of 10 scenarios
    0.956
    • FastAPI 0.111.0
    • ChromaDB
    • LangChain

    FastAPI backend with ChromaDB and LangChain embedding and chat clients (committed code at HEAD).

    No automated tests.

    Source: LexIntelligence (opens an external site)

  • PhishSniffer

    Phishing-email classifier with a Streamlit UI, built as a Telecom Egypt internship deliverable.

    Test accuracy8,571 held-out samples, committed model metadata
    97.7 %
    • scikit-learn
    • Streamlit
    • Python

    scikit-learn models: Random Forest, Gradient Boosting, Logistic Regression, on TF-IDF plus handcrafted features.

    Trained on about 43K emails from public corpora (34,284 train, 8,571 held out).

    Source: PhishSniffer (opens an external site)Live demo: PhishSniffer (opens an external site)May take a few seconds to wake.

03 / Open source

Open source

Status from GitHub, refreshed daily · Aug 2026 – Oct 2026

Issues reported

04 / Experience

Experience

  1. –

    Systems Engineering Intern

    Alstom

    • Cairo Monorail (CITYFLO 650 CBTC): diagnosed in-service faults on the driverless line by tracing 10+ subsystem interfaces (ATS, interlocking, ATP/ATO, platform screen doors, SCADA) and rebuilding event timelines from onboard and wayside logs to separate root causes from symptoms.
    • Audited 660 pages of wayside and onboard maintenance manuals against the as-built architecture: a register of 78 findings (11 critical), each checked against its source document through 150+ scripted text checks.
    • AI/ML sprint on railway incident data: framed the problem of 5 incompatible alarm formats feeding one operator screen, set 5 data-quality metrics, and proposed a unified incident taxonomy and classification approach.

  2. IT Infrastructure Intern

    Orascom Construction PLC

    • CCNA-aligned networking: routing, VLAN segmentation, NAT, IP addressing, switching.
    • Guided data-center rack walkthrough: firewall, servers, UPS.

  3. –

    AI/ML Intern

    Telecom Egypt (WE)

    • Delivered 2 AI-enhanced data-processing prototypes with supervised and unsupervised ML; one was PhishSniffer.
    • EDA, preprocessing, and evaluation with precision, recall, F1 and cross-validation.

    Project: PhishSniffer (opens an external site)

Education

  • B.Sc. Software Engineering

    Arab Academy for Science, Technology and Maritime Transport (AASTMT)

    –

Certifications

  • Claude 101, Claude Code 101, Claude Code in Action, Agent Skills, Subagents

    Anthropic Academy

  • GenAI Practice

    NVIDIA

  • Building LLM Applications with Prompt Engineering

    NVIDIA

  • AI Associate

    Telecom Egypt

  • DevOps Foundations

    Sprints

Every figure has a source.

Automated tests
228
Eventora · 228 / 228 passing
Event-Ticketing-Platform/README.md:253 (opens an external site) · as of
JaCoCo instruction coverage
84.1 %
Eventora · gated packages, gate 80%
Event-Ticketing-Platform/README.md:254 (opens an external site) · as of
Booking states
10
Eventora
Event-Ticketing-Platform/src/main/java/com/ticketing/booking/model/BookingState.java (opens an external site) · as of
Flyway migrations
14
Eventora
Event-Ticketing-Platform/src/main/resources/db/migration (opens an external site) · as of
Branch coverage
84.07 %
Recruiter-Pro · CI floor 81%
Recruiter-Pro/README.md:37; .github/workflows/ci.yml:86 (opens an external site) · as of
Canonical skills
679
Recruiter-Pro
Recruiter-Pro/README.md:49 (opens an external site) · as of
Architecture decision records
3
Recruiter-Pro
Recruiter-Pro/docs/adr (opens an external site) · as of
merged pull requests
15
11 projects, 3 under review
GitHub API, gh pr view per pull request · as of
pytest tests
104
SysPlex
evidence-report.md:143 · as of

Systems that keep working as load grows.

Send a message

05 / Contact

Contact

Send a message. It goes to my inbox, and I reply by email.

10 to 4000 characters.