CYBERSECURITY // AI/ML // NETWORKING ENGINEER

Hi, I'm
Amna Shabir

Results-driven IT & Cybersecurity professional with 2.5+ years of hands-on experience in penetration testing, network security, and AI/ML development. I build recon frameworks, AI-driven security proofs-of-concept, and offensive security tooling for real-world experimentation.

OFFENSIVE SECURITY AI / ML PENETRATION TESTING PYTHON / KALI LINUX DIGITAL FORENCICS GOOGLE CYBER CERT
VIEW PROJECTS GET IN TOUCH
// ABOUT

I'm a cybersecurity and AI/ML professional with 2.5+ years of experience, building offensive security tooling, reverse engineering solutions, and hands-on experimentation labs for real threats.

My work spans CLI-based vulnerability tools, AI-powered automation systems, network configuration, and IoT analysis. I hold a Google Cybersecurity Certificate and have participated in national digital forensics challenges.

Currently pursuing a B.S. in Information Technology at Shaheed Benazir Bhutto University (2024–2028).

SKILLS // STACK
Ghidra / IOT
Penetration Testing
Linux / Bash
NLP / LLMs
OWASP
Python
Machine Learning
SQL
SIEM Tools
Cisco Networking
// PROJECTS
FEATURED PROJECT // OFFENSIVE TOOLING
ReconKit — Concurrent Reconnaissance Framework

Problem: Security reconnaissance tools are often fragmented, slow, or difficult to extend for large-scale analysis.

Approach:
• Designed a modular reconnaissance framework in Python (~1500 lines)
• Implemented concurrent scanning for significantly improved performance
• Integrated multiple reconnaissance techniques into a unified and extensible pipeline
• Containerized the solution using Docker for easy deployment and reproducibility

Key Features:
• Parallelized scanning workflows
• Modular architecture supporting rapid extension and adaptation
• Automated aggregation and reporting of recon data for structured analysis

Tools: Python, Docker

Outcome: Developed a scalable and extensible reconnaissance system suitable for structured security analysis and experimentation, greatly reducing manual effort and increasing coverage for pentesters and researchers.

At a Glance
  • Unified pipeline: Whois, DNS, subdomain discovery, port scan, fingerprinting
  • Concurrent execution dramatically speeds up reconnaissance phases
  • Fully automated HTML/TXT reporting
  • Extensible: new recon modules can be integrated with minimal effort
PYTHON DOCKER PARALLELISM MODULARITY SECURITY AUTOMATION
02 // OFFENSIVE SECURITY
PayloadForge — Educational Offensive Security Framework

Problem: Understanding vulnerabilities requires hands-on exposure to how exploits are constructed and executed.

Approach:
• Built an OWASP-aligned framework demonstrating real-world exploitation logic:
   – Cross-Site Scripting (XSS)
   – SQL Injection
   – Command Injection
• Designed the system for reproducibility and controlled, safe testing environments
• Targeted for educational and research use, prioritizing understanding over automation

Key Contributions:
• Structured exploit-generation workflows reflect real-world attack chains
• Demonstrated common pitfalls and how input validation fails
• Provided a safe sandbox for vulnerability experimentation and training

Tools: Python, Web Security Tools

Outcome: Enabled systematic exploration of common web vulnerabilities and their root causes, delivering value to students and security enthusiasts learning about exploitation.

PYTHON XSS SQLi CMDi EDUCATION OWASP
03 // RESEARCH
Cloud-Native Microservices Vulnerability Analysis (Research Work)

Problem: Modern microservice architectures introduce complex attack surfaces that traditional scanners and analysis tools frequently miss.

Approach:
• Analyzed the open-source OWASP crAPI deployed within a Kubernetes environment
• Applied static analysis (Ghidra), traffic inspection (Wireshark), and API testing (Burp Suite)
• Reconstructed inter-service logic and behavioral flows, mapping out hidden trust boundaries

Key Findings:
• Exposed business logic inside internal microservices invisible to black-box testers
• Identified cross-service authentication bypass and orchestration permission flaws
• Surfaced vulnerabilities requiring complex cross-layer analysis, beyond the reach of conventional tools

Outcome: Advanced cloud security research, highlighting the need for multi-layer visibility in modern app assessments and uncovering new vulnerability classes in production-grade microservices.

GHIDRA KUBERNETES WIRESHARK BURP SUITE MICROSERVICES CLOUD SECURITY
04 // HARDWARE / RESEARCH
IoT Attack & Defense Toolkit

An end-to-end IoT attack & defense research laboratory on ESP32 — includes firmware reverse engineering, protocol exploitation, and automated threat discovery. Demonstrated MQTT vulnerabilities, performed binary analysis (Ghidra), simulated attacks, and generated CVE-style disclosures for embedded hardware.

Key Highlights: Custom exploit firmware, binary extractions using esptool.py, traffic capture with Wireshark, and automated reporting for IoT vulnerability research.

ESP32 PYTHON REVERSE ENGINEERING MQTT GHIDRA WIRESHARK
// PUBLICATIONS
“Unveiling the Hidden Attack Surface: Reverse Engineering Techniques Applied to Cloud-Native Microservices”
Presented at IC-ESGFCR 2026 International Conference
// EXPERIENCE
Offensive Security Intern
JAN 2026 — APRIl 2026
ITSOLERA PVT LTD // ISLAMABAD, PAKISTAN
  • Built ReconKit — modular CLI recon framework cutting manual recon time by 70%
  • Engineered PayloadForge — educational payload generator with XSS, SQLi, and CMDi modules aligned with OWASP standards
  • Integrated multi-encoding support (URL, Base64, Hex) and obfuscation logic into payload templates
AI / ML Intern
SEP 2025 — NOV 2025
ZENITHCRAFT VENTURES (PVT) LTD // LAHORE, PAKISTAN
  • Engineered AI story teller using two LLMs and prompt templates, reducing manual effort by 40%
  • Collaborated on live AI/ML model development, solving data processing issues to boost accuracy
  • Researched AI hallucinations and applied prompt engineering to ensure ethical, reliable outputs
IT & Software Intern
JUL 2025 — AUG 2025
SUKKUR BEVERAGES PVT. LTD. — PEPSICO FRANCHISE // SUKKUR, PAKISTAN
  • Built AI chatbot for sales team using NLP APIs, improving query resolution by 25%
  • Configured VLANs and remote access in Cisco Packet Tracer for network security prep
  • Led team to 1st place and Best Project Award in IT competition
  • Earned Participation Award for network maintenance and departmental contributions
// CONTACT

Let's work together

Open to security roles, AI/ML projects, and freelance pentesting engagements.