I build secure, explainable systems that turn complex data into useful intelligence.
I'm an Information and Communication Technology Engineering student at Metropolia University of Applied Sciences in Finland. My interests sit at the intersection of cybersecurity, data engineering, artificial intelligence and practical software development.
- 🔭 Building privacy-conscious CTI, OSINT and data-analysis systems
- 🧠 Interested in explainable AI, local inference, LLM agents and vector memory
- 🛡️ Experienced with IOC analysis, log analysis, static malware analysis and digital forensics
- ⚙️ Comfortable across backend, web, desktop, databases, Linux and containerized environments
- 🌱 Always learning through real, end-to-end projects
- 🤝 Open to remote internships and meaningful technical collaborations
- 🌍 Languages: Turkish, Finnish and English
Private, explainable cyber-threat-intelligence search for an authenticated X timeline
Timeline CTI Explorer is an end-to-end intelligence platform that collects posts from an operator's own authenticated timeline, extracts threat indicators, assigns an explainable CTI score and makes the enriched data searchable through a protected analyst console.
X Timeline → Encrypted SQLite WAL → CTI Enrichment → ClickHouse → FastAPI → React Console
Key engineering highlights:
- Dual ingestion: Selenium-based collection or the official X API with OAuth 2.0 + PKCE
- Threat enrichment: IPv4/IPv6, domains, URLs, hashes, CVEs, MITRE ATT&CK techniques and actor identifiers
- Explainable scoring: deterministic rules combined with optional local multilingual ONNX inference
- Resilient pipeline: AES-256-GCM encrypted state, durable WAL spool, deduplication, retry and audit trails
- Fast investigation: ClickHouse native text indexing with full-text, IOC and CTI filters
- Defense in depth: Argon2id, CSRF protection, strict cookies, isolated networks, least-privilege containers and HTTPS
- Operational visibility: structured logs plus optional Prometheus and Grafana services
The project reflects how I like to work: privacy first, measurable behavior, explicit trade-offs and documentation that treats security as part of the architecture.
Languages & application development
Python · C++ · C · Bash · JavaScript · TypeScript · FastAPI · Flask · React · Qt · PyQt
Data, AI & computer vision
pandas · PyTorch · ONNX Runtime · scikit-learn · OpenCV · YOLO · InsightFace · Model fine-tuning · LLM agents
Databases & infrastructure
ClickHouse · PostgreSQL · MySQL · SQLite · DuckDB · Milvus · Docker · Docker Compose · Linux
Cyber intelligence & forensics
CTI · OSINT · Dark-web intelligence · IOC analysis · Static malware analysis · Log analysis · Digital forensics
I want to contribute to teams where cybersecurity, AI and data engineering solve real problems—not just produce demos. I am especially interested in opportunities that involve:
- threat intelligence and security analytics;
- privacy-preserving AI and local inference;
- search, enrichment and high-volume data pipelines;
- computer vision and intelligent automation;
- backend or full-stack product engineering.
Türkçe kısa profil
Finlandiya'da Metropolia Uygulamalı Bilimler Üniversitesi'nde Bilgi ve İletişim Teknolojileri Mühendisliği öğrencisiyim. Siber tehdit istihbaratı, veri mühendisliği, yapay zekâ ve yazılım geliştirme alanlarına odaklanıyorum. Güvenli, açıklanabilir ve gerçek dünyada kullanılabilir sistemler geliştirmeyi seviyorum. Uzaktan staj ve teknik iş birliği fırsatlarına açığım.
If you're working on a challenging problem in cybersecurity, data or AI, I'd be happy to hear from you.
Curious by nature. Security-minded by design. Always building.