Available for SOC & Cybersecurity Analyst roles · Charlotte, NC · Remote OK

Michael Kurdi

/ SOC-oriented security analyst · detection engineer · builds Python tooling & LLM expert systems.

michael@kurdi: ~ zsh

michael@kurdi :~$ whoami

8+yrs cleared ops
4shipped security builds
99.68%phishing F1 (5-fold CV)
3.96GPA · summa cum laude

01 / about

$ cat about.md

I'm Michael Kurdi — a CompTIA Security+ professional with more than eight years in federally regulated, security-cleared environments (FAA, OSHA, IATA, CBP). I hold a B.S. in Information Technology (Cybersecurity) from SNHU (Summa Cum Laude, 3.96 GPA) and focus on incident response, detection engineering, and Python security tooling.

I'm targeting SOC Analyst / Cybersecurity Analyst roles and I ship real code: rule-based log anomaly engines, ML phishing classifiers with cross-validated metrics, and prompt-engineered expert systems packaged with docs, tests, and print-ready resumes. I care about reproducibility, clean READMEs, and outputs a human reviewer can actually verify.

Press ⌘K anywhere to open the command palette. Or jump to #terminal for an interactive shell.

~/whoami.json

{
  "name": "Mahmoud (Michael) Kurdi",
  "role": "SOC / Cybersecurity Analyst",
  "location": "Charlotte, NC",
  "remote": true,
  "cert": "CompTIA Security+ ce",
  "stack": ["python", "flask", "react", "sklearn"],
  "currently": "building detection tooling & LLM eval harnesses",
  "open_to": ["SOC", "SecEng", "Detection"]
}

Detection & analytics

Rule-based and statistical engines for logs and email. Anomaly categories, SOC-style triage output, ML pipelines with measured cross-validated performance.

Expert LLM systems

Role priming, XML structuring, CoT, ReAct, self-reflection, and hallucination guardrails — shipped as reproducible prompts with GitHub-ready docs and UIs.

Full-stack delivery

Python, React, Flask, SQLite, REST APIs, CI-friendly static sites on GitHub Pages — with READMEs and UX that hold up to real reviewers.

02 / projects

$ ls -la ./featured

Open source, documented, deployable. No hand-waving — every repo has a README and runnable code.

Prompt System · React featured

StockPath Navigator ↗

AI-powered trading expert system · 17 prompt-engineering techniques

stockpath-navigator

// analyze AAPL — ReAct loop

user > analyze AAPL trend

◆ reasoning: bullish pattern

◆ risk: moderate

◆ confidence: 87%

⚡ safety guardrails active

A disciplined system prompt that turns any reasoning-capable LLM into a stock market analyst and trading coach. Real-time analysis, trade journaling, emotional-state management, and an interactive React dashboard.

Prompt Engineering
ReAct
CoT
React
Safety Guardrails

Python · Flask · SQLite v1.0

VulnTrack Dashboard ↗

Full-stack vulnerability management · KPIs · CRUD · REST

vulntrack · GET /api/vulnerabilities

// severity triage dashboard

200 OK 12 records loaded

critical: 3

high: 5

resolved: 4 patched

Full-stack dashboard with KPI tracking, color-coded severity, CRUD, live search/filter, and toast notifications — pre-seeded with 12 sample records. Dark-mode cybersecurity branding on a REST API.

Python
Flask
SQLite
REST
CRUD

Python · scikit-learn ML

Phishing Email Classifier ↗

TF-IDF + Random Forest · 99.68% F1 · 5-fold CV · interpretable

phishing-email-classifier

# RandomForest + TF-IDF pipeline

>>> features_engineered = 207

>>> f1_cv5 = 0.9968

>>> importances.head(5)

# surfaces what drove each label

Text-based feature extraction with TF-IDF plus custom regex signals, feature-importance analysis for interpretability, and SOC-style reasoning about what drove each label.

Python
scikit-learn
TF-IDF
Cross-validation

Python · pandas · NumPy SOC

Log Anomaly Engine ↗

Rules + statistics for brute force, port scans, volume spikes

detect --input auth.log

# rule + statistical engine

brute_force threshold hit (5/60s)

port_scan pattern detected

z_score volume spike (3.4σ)

alerts structured for SOC triage

Brute-force thresholds, port-scan patterns, off-hours access, unknown-IP heuristics, privilege escalation, and z-score volume spikes — with output structured for SOC-style triage pipelines.

Python
pandas
NumPy
Log analysis

03 / stack

$ cat ~/.stackrc

Daily drivers — tools I actually ship with, not buzzwords.

Languages

Python90%
JavaScript / TypeScript75%
Bash / Shell70%
SQL65%

Security

Detection engineering85%
Incident response80%
Vulnerability mgmt75%
Access control / IAM70%

Frameworks & Data

Flask · FastAPI80%
React75%
pandas · NumPy85%
scikit-learn78%

Tooling

Git · GitHub Actions90%
VS Code · Cursor85%
Docker70%
LLM APIs · Prompt Eng80%

pythonreactflask fastapipandasnumpy scikit-learntf-idfrandom-forest sqliterestgit dockergithub-actionsbash regexprompt-engineeringreact-native socsiemincident-response siem-huntingmitre-att&ckyara

04 / timeline

$ git log --oneline --career

Commits to the career repo. Most recent first.

a1b2c3d

2026HEAD → main

Shipped StockPath Navigator v3.0 + portfolio overhaul

17-technique prompt-engineered trading system with React UI. Refreshed this site for a developer-first feel — keyboard shortcuts, interactive terminal, SEO polish.
f4e5d6a

2025milestone

Earned B.S. IT Cybersecurity · summa cum laude · SNHU

3.96 GPA. President's List. Alpha Sigma Lambda. Capstone in detection/response fundamentals.
9c8b7a6

2025cert

Passed CompTIA Security+ ce

Foundational certification aligned with SOC and GRC expectations.
5d4c3b2

2024build

Open-sourced phishing classifier + log anomaly engine

ML and rule-based detection with cross-validated metrics, feature interpretability, and SOC-shaped triage output.
1a2b3c4

2017–presentops

8+ yrs · cleared, regulated operations

FAA, OSHA, IATA, CBP compliance. Security clearance + CBP badge endorsement. Access control, audit readiness, incident-handling reflexes you can't fake.

05 / terminal

$ ./portfolio.sh

A real sandboxed shell. Try help, about, projects, skills, contact, sudo hire-me.

zsh — michael@kurdi — 120×40

06 / contact

$ curl michael --hire

Open to SOC Analyst / Cybersecurity Analyst / Detection Engineering · remote or Charlotte, NC.

Email me Download resume.pdf Web resume

GitHub

@KM-it-ops

michael-kurdi

kurdi.michael.it@gmail.com