Skip to content

Chirra Karthik

Cybersecurity Engineer • c***************@gmail.com • +91*******613 • drivetube.ai/•••••

Career Objective

Cybersecurity Engineer with 0 years of experience focused on network security, encrypted traffic analysis, and ML-driven intrusion detection. M.Tech in Computer Networks and Information Security (JNTUH) with hands-on academic projects building hybrid ML/DL models for real-time encrypted traffic classification, explainable IDS using UNSW-NB15, and embedded smart-home solutions on Raspberry Pi. Seeking an entry-level role applying ML and network security skills to detect and mitigate threats while preserving user privacy.

Education

J. N. T. University Hyderabad (JNTUH UCESTH)
M.Tech. in Computer Networks and Information Security (CSE) • 2024 – 2026
Institute of Aeronautical Engineering (JNTUH)
B.Tech. in Electronics and Communication Engineering • 2019 – 2023
Narayana Junior College
Class 12 (P.C.M.) • 2017 – 2019
Sharada Vidya Bhavan High School
Class 10 • 2017

Technical Skills

Programming Languages: Python,MATLAB
Web Technologies: HTML
Frameworks and Libraries: scikit-learn,TensorFlow,Keras,PyTorch,pandas,numpy,OpenCV
Databases: SQL,DBMS,SQL querying
Cloud and DevOps: Linux
Data and Analytics: ANN,Power BI,data visualization
Tools and Methodologies: Jupyter,Google Colab,Visual Studio,Raspberry Pi,Git
Explainability & ML Ops: SHAP,LIME,model evaluation,data preprocessing
Cybersecurity & Networking: Network Traffic Analysis,Encrypted Traffic Classification,Intrusion Detection Systems,TCP,IP,Network Security,UNSW-NB15 dataset
Operating Systems: Windows,Android

Projects

IntelliSec: An Hybrid Adaptive AI Framework For Real-Time Encrypted Traffic Classification
Tools Used: Encrypted Traffic Classification, Machine Learning, Deep Learning, Python, scikit-learn, TensorFlow, Network Traffic Analysis
  • Designed and implemented a hybrid ML/DL framework to classify encrypted network traffic in real time without decrypting payloads, preserving user privacy.
  • Engineered feature extraction from flow-level network metadata and protocol statistics to enable payload-agnostic classification.
  • Combined classical ML models and deep learning modules to improve robustness across diverse traffic types and achieve 98.7% classification accuracy on test sets.
  • Optimized inference pipeline for low-latency operation suitable for inline or edge deployment using model pruning and batch sizing.
  • Validated model performance across encrypted protocols and measured precision/recall to ensure low false-positive rates for security monitoring.
Advancing Cyber Security Using Explainable Artificial Intelligence
Tools Used: Intrusion Detection Systems, UNSW-NB15, ANN, SHAP, LIME, Python
  • Developed an ML-based intrusion detection system using the UNSW-NB15 dataset to identify cyberattack patterns in network flows.
  • Trained artificial neural network classifiers and evaluated using accuracy, precision, recall, and F1 metrics to assess detection capability.
  • Integrated SHAP and LIME explainability to provide transparent, instance-level explanations for model alerts to support analyst triage.
  • Implemented data preprocessing, feature engineering, and class imbalance handling to improve detection of subtle attack types.
  • Produced visualization dashboards to present explainability outputs and detection trends for non-technical stakeholders.
Smart Home Virtual Assistant Using Raspberry Pi
Tools Used: Raspberry Pi, Embedded Systems, Face Recognition, OpenCV, Python
  • Designed a smart-home virtual assistant prototype on Raspberry Pi integrating services for security and entertainment.
  • Implemented face-recognition based access control using OpenCV and lightweight CNN models to enable local authentication.
  • Developed modules for voice interaction and device control to unify smart-home services beyond mobile-only assistants.
  • Integrated local logging and alerting to support surveillance and security use-cases in an offline environment.
  • Conducted usability testing to refine voice-command flows and recognition reliability under varied lighting and noise.
Bringing Old Photos Back To Life (Image Restoration & Denoising)
Tools Used: Image Restoration, Denoising, Deep Learning, Python, MATLAB, OpenCV
  • Researched and implemented restoration techniques to reconstruct degraded old photographs with mixed structured and unstructured defects.
  • Applied deep learning-based denoising models to increase contrast and recover lost detail in low-SNR cryo-TEM style images.
  • Applied hybrid pipelines combining classical image processing with learned priors to handle varied degradation patterns.
  • Evaluated visual quality improvements using structural similarity and qualitative expert review to validate restoration effectiveness.
  • Optimized preprocessing and augmentation to improve model generalization on diverse historical photo datasets.

Certifications

Accenture - Forage Certification, Data Analytics & Visualization of Virtual Experience — Accenture (Forage) • 2023
Intro to Data Science — Coursera • 2021

Achievements

  • Research Publication - INTELLISEC: An Hybrid Adaptive AI Framework For Real-Time Encrypted Traffic Classification — June-2026: Published in IJEDR - International Journal of Engineering Development and Research, Vol.14, Issue 2, pages 258-267.
  • Organizing Secretary, Lexicon 2.0, IEEE-IARE — December 18-19, 2021: Coordinated event logistics and technical sessions for student-run technical symposium.
  • Secretary, Street Cause, IARE Cell — Apr. 2019 - Apr. 2020: Led community engagement and outreach initiatives.
  • Coordinator, Code Quest, Spirit JNTUH — April 15-16, 2025: Managed coding competition organization and participant coordination.

Powered by Drivetube · Create your own profile at drivetube.ai