> Initializing system...
> Loading portfolio components...
> Establishing secure connection...
> Welcome to the portfolio of akhilvvn
Choose your preferred interface experience
Building intelligent AI solutions for cybersecurity with expertise in threat detection, computer vision, and explainable AI. Creating secure, privacy-focused machine learning systems for real-world security applications.
Get to know me better
I am an AI/ML Engineer specializing in Cybersecurity with a unique background in Criminology and Police Administration. This distinctive combination allows me to approach security challenges from both technical and investigative perspectives, building AI-powered solutions that address real-world threats.
As the HAC'KP 2025 Winner, I built the age estimation module for the Kerala Police CyberDome take-it-down platform, enhancing law enforcement capabilities with computer vision models optimized for real-time operational environments. My work focuses on security-focused AI applications including threat detection systems, forensics tools, and privacy-preserving machine learning.
My technical expertise spans explainable AI, generative models, and full-stack development, enabling me to create end-to-end ML pipelines that are not just functional but fundamentally secure by design. I combine offensive security knowledge with cutting-edge AI/ML techniques to build robust defense systems and investigative tools.
Security AI Specialist
Deep Learning Engineer
CV Systems Developer
What I bring to the table
Some of my recent work
Built hybrid ML system with XGBoost and Autoencoders combined with LLM reasoning for real-time threat detection. Implemented 5-stage Google Gemini pipeline with SHAP explainability and interactive visualizations achieving sub-3s WebSocket latency.
Developed 8-tool AI/ML suite for image forensics including perceptual hashing, ResNet scene classification, CLIP-based multimodal retrieval, YOLOv8 object detection, and FastAPI + React investigator tool for comprehensive image analysis.
Built Random Forest model with SHAP explainability and gender fairness analysis. Deployed via Streamlit with interactive visualizations for live predictions and feature impact analysis, achieving strong performance metrics.
Engineered client-side age estimation using faceapi.js and TensorFlow.js with on-device inference. Optimized lightweight models for low-latency predictions in production environments without compromising user privacy.
My professional journey
Kerala Police Cyberdome
Edunet Foundation (AICTE & IBM SkillsBuild) • June 2025 – July 2025
Industry-recognized credentials
Let's work together
Thiruvananthapuram, Kerala, India
Currently available for AI/ML and cybersecurity internships, full-time roles, research collaborations, and freelance projects.