| # -------- charan@ankamteja ---------OS: ......... Kali, Fedora 43, Win 11Uptime: ............ 19 y, 5 mo, 14 dHost: ............ Amrita Vishwa, CBEKernel: .......... Offensive SecurityShell: .................... bash, zshIDE: ................... VS Code, vimLang.Code: ........ Python, C++, JavaLang.Web: ........... HTML, CSS, YAMLLang.Real: ... English, Telugu, HindiHobbies: ....... CTFs, HTB, Write-UpsLearning: ....... DSA, ML, NetworkingCerts.WIP: ........... HTB CPTS, CCNAToolkit: ....... Burp, Wireshark, MSF# ------------- Contact -------------Email: ... ankamcharantejaa@gmail.comLinkedIn: ........ in/ankamcharantejaHackTheBox: ............... ankamtejaMedium: ........... @ankamcharantejaaGitBook: ........ capcap-1.gitbook.io# ---------- GitHub Stats -----------Repos: .......... 17 owned | 2 forkedDisclosure: ...... CERT-In / 74835826 |
| Repository | Focus & Technical Stack |
|---|---|
| Vulnerability-Assessment-Case-Study | Independent, passive security assessment of a public-sector web application. 6 vulnerabilities identified (1 Critical, 3 High, 2 Medium) through OSINT and enumeration only — no exploitation — and responsibly disclosed to CERT-In (CERTIn-74835826). |
| air-gapped-mpls-copilot | Project Aether — an offline, air-gapped AI copilot for MPLS network operations with zero cloud dependency. Pairs a stochastic LSTM (anomaly + fault classification) with a deterministic NetworkX graph model, and only acts when both corroborate. Containerlab 7-node MPLS L3VPN simulation, Prometheus/Grafana telemetry, and an offline LLM for plain-English explanations. Python. |
| pe-malware-detection | Binary Benign vs Malware classification of Windows PE executables from 77 static header and import-table features (21.7k de-duplicated samples across 26 malware families). Random Forest with SMOTE balancing — 0.992 F1 on a held-out test set — evaluated with confusion matrix, ROC/PR curves, and feature importance. Jupyter. |
| CyberStegoApp | Android steganography app that hides messages inside images via LSB. Payload sealed with authenticated AES-256-GCM (PBKDF2 key derivation, per-message salt + IV, tamper detection); account passwords stored as salted PBKDF2 hashes. Room session store, shake-to-logout. Java. |
| Repository | Focus & Technical Stack |
|---|---|
| ad-dmsa-privilege-escalation-lab | Detection lab for the "BadSuccessor" dMSA privilege-escalation family in Windows Server 2025 Active Directory. Reproduces the attack chain in an isolated lab and builds the missing detection content — Sigma / Wazuh / Splunk rules plus a PowerShell exposure-enumeration script. Defensive research, built jointly with @Hemashankar19. |
| turbojet-digital-twin | Physics-informed digital twin for a four-stage turbojet — estimates hidden component health and performance (thrust, TSFC) from limited sensor data using physics-constrained surrogate models, with an interactive 3D dashboard. Python · React/Three.js. |
| latency-aware-edge-game-placement | Systems research measuring how real server load breaks the analytical latency models used to place edge game servers — quantifying where the theory diverges from measured behaviour. JavaScript. |
More:Write-Ups - CTF & security writeups.
- Technical Articles:medium.com/@ankamcharantejaa
- Documentation:capcap-1.gitbook.io/capcap
