I build, investigate, and secure digital systems.

I am a cybersecurity student focused on understanding how systems fail, how vulnerabilities can be identified, and how practical security solutions can be designed and tested.

Portrait of Prajwal K

Curiosity is where security engineering begins.

I study cybersecurity with a focus on cryptographic protocols, threat detection systems, and systems-level security. My work spans both offensive security — reverse engineering, binary analysis, and CTF challenges — and defensive engineering: building autonomous intrusion prevention systems, implementing post-quantum cryptographic primitives, and designing secure communication protocols.

I approach each project by understanding the underlying system first, identifying where it can fail, and then building or testing solutions that address those specific failure modes. I am looking for internship and entry-level opportunities in cybersecurity engineering, application security, security research, or systems engineering.

“Engineering defense requires first mastering where, why, and how complex software abstractions fail.”
Current Focus Areas06 DOMAINS
Post-Quantum Cryptography01
Intrusion Detection & Prevention02
Reverse Engineering03
Application Security04
Network Security05
Secure Systems Design06
Target: Internship & Security Engineering Opportunities (2027)

Engineered systems & security architectures.

01/Post-Quantum Cryptography
Completed

QUANTA↗

PROBLEM THESIS:

Modern public-key cryptographic infrastructure (RSA, ECDH, ECDSA) relies on mathematical problems like integer factorization and discrete logarithms that will be solved in polynomial time by Shor's algorithm on cryptanalytically relevant quantum computers (CRQCs). Hostile entities are actively executing 'Harvest-Now, Decrypt-Later' (HNDL) campaigns, intercepting and storing encrypted network sessions today to decrypt once quantum computing hardware matures. Existing application architectures lack standardized post-quantum key encapsulation mechanisms (KEM) and authenticated post-quantum digital signatures.

TECHNICAL APPROACH:

Implemented a comprehensive post-quantum cryptographic protocol following NIST FIPS 203 (ML-KEM-768) and FIPS 204 (ML-DSA-65) lattice standards. Built an authenticated key exchange protocol combining quantum-resilient key encapsulation with classical defense-in-depth via HKDF-SHA256 hybrid key derivation. The derived symmetrical session keys feed AES-256-GCM AEAD encryption engines for tamper-resistant data transit. Accompanying the protocol is a real-time cryptographic telemetry dashboard for tracking encapsulation cycle latency, public key footprint overheads, and handshake verification metrics.

SYSTEM STACK
Java CryptographyML-KEM-768 (NIST FIPS 203)ML-DSA-65 (NIST FIPS 204)AES-256-GCM
Case Study→|Source ↗
02/Intrusion Detection & Prevention
Completed

SSH BruteForce Detector↗

PROBLEM THESIS:

Internet-facing Linux hosts endure relentless SSH credential stuffing and dictionary attacks by distributed botnets. Relying solely on manual syslog review or delayed log processors leads to multi-hour detection windows during which compromise can occur. Traditional static firewall configurations cannot dynamically distinguish between legitimate authentication spikes and multi-threaded credential exhaustion campaigns.

TECHNICAL APPROACH:

Engineered an autonomous host-based intrusion prevention system (HIDS/IPS) in Python. The daemon asynchronously tails /var/log/auth.log via kernel inotify events, parsing authentication failures in real time. It implements an in-memory sliding-window rate tracking algorithm that aggregates failure events by source IP. Upon threshold breach, the system executes kernel-level Netfilter/iptables DROP commands with automated expiration and ban lifecycles. A Flask-powered SOC monitoring dashboard visualizes blocked subnets, attempt velocity, and ban table telemetry.

SYSTEM STACK
Python 3Linux iptables / NetfilterFlaskLinux Kernel Inotify
Case Study→|Source ↗
03/Secure Communications
Completed

Entangle↗

PROBLEM THESIS:

Contemporary mobile messaging applications present opaque cryptographic implementations or store sensitive metadata in proprietary cloud infrastructure. Furthermore, standard end-to-end encrypted messaging systems rely exclusively on classical elliptic curve cryptography (Curve25519 / NIST P-256), rendering all stored encrypted communications permanently decipherable once quantum computers achieve cryptanalytic scale.

TECHNICAL APPROACH:

Engineered an Android end-to-end encrypted messaging platform using Kotlin and Jetpack Compose. Built a client-side cryptographic architecture where key generation, key agreement, and payload encryption occur strictly on the user's mobile device. Implemented X25519 Elliptic Curve Diffie-Hellman (ECDH) key exchange augmented with post-quantum key agreement primitives. Encrypted message bodies are sealed using AES-256-GCM authenticated cipher with unique initialization vectors per message. Cloud relays (Firebase Firestore) operate purely as blind ciphertext stores with zero visibility into plaintext content.

SYSTEM STACK
KotlinAndroid SDK / Jetpack ComposeX25519 ECDHAES-256-GCM AEAD
Case Study→|Source ↗
04/Machine Learning & Security Analytics
Completed

Multimodal Emotion & Engagement↗

PROBLEM THESIS:

Single-modality emotion and sentiment recognition systems suffer from high variance and deception susceptibility. Facial expression classifiers fail under variable illumination, head pose changes, and voluntary expression suppression. Conversely, text-only sentiment models miss sarcasm, stress cues, and emotional intensity. Combining heterogeneous modalities (visual facial embeddings and natural language text tokens) requires robust feature normalization and low-latency inference fusion.

TECHNICAL APPROACH:

Developed a multimodal engagement analytics platform fusing real-time computer vision with transformer-based NLP. The visual pipeline employs an OpenCV face detection stage feeding a convolutional neural network (CNN) trained on the FER2013 dataset. The text pipeline processes linguistic input using a fine-tuned Hugging Face DistilRoBERTa model. Extracted feature embeddings from both models are normalized into a unified 14-dimensional feature vector and fused using a late-fusion K-Nearest Neighbors (KNN) classifier. The complete inference pipeline is exposed via high-throughput FastAPI endpoints.

SYSTEM STACK
Python 3OpenCVKeras / TensorFlowHugging Face DistilRoBERTa
Case Study→|Source ↗
05/Anonymous Communications, Host Privacy & Web3 Micropayments
Completed

AfterClass↗

PROBLEM THESIS:

Traditional messaging networks demand invasive personal identifiers (phone numbers, emails, social logins), persist invasive device telemetry, and store user metadata in centralized databases susceptible to leaks and surveillance. Furthermore, monetizing or supporting creators on anonymous platforms has historically required fiat payment rails (credit cards, PayPal) that completely destroy pseudonymity via mandatory KYC. Self-hosted message boards also expose the host's residential IP address, require complex router port-forwarding, suffer from database write-lock contention under concurrent room traffic, and leak geographic coordinates through unstripped image EXIF metadata.

TECHNICAL APPROACH:

Engineered an asynchronous real-time message exchange using FastAPI and Starlette WebSockets backed by SQLite in Write-Ahead Logging (WAL) mode, combined with a comprehensive Ethereum Sepolia / Base Sepolia Web3 micropayment architecture. Authored and deployed custom Solidity smart contracts: the AfterClass ERC-20 utility token ($AFTER) featuring an automated cooldown faucet and on-chain post tipping events, alongside PremiumStore.sol supporting EIP-2612 gasless permit signatures (ECDSA r, s, v) so users can purchase cosmetic VIP badges and spotlight glows without paying gas fees. Built an asynchronous Web3.py backend service with multi-provider RPC failover (Infura, Tenderly, dRPC, Ankr) for non-blocking transaction confirmation, coupled with automated Cloudflare Tunnel daemon routing, in-memory Pillow EXIF stripping, and 256-bit cookie token host administration.

SYSTEM STACK
FastAPI (Python 3.12)Async WebSockets (Starlette)Solidity (ERC-20 / EIP-2612)Ethereum Sepolia & Web3.py
Case Study→|Source ↗
ALL SECURITY IMPLEMENTATIONS ACCOMPANIED BY BENCHMARK TELEMETRY
Explore all repositories on GitHub ↗

Tools are useful. Understanding matters more.

CATEGORY / 01DOMAIN

Programming & Scripting

Systems programming, security tooling, and automation.

  • Python
  • C / C++
  • Kotlin
  • Bash
  • SQL
  • JavaScript
CATEGORY / 02DOMAIN

Security & Forensics

Offensive testing, traffic analysis, and binary examination.

  • Reverse Engineering (Ghidra, GDB)
  • Network Analysis (Wireshark, tcpdump)
  • Burp Suite
  • Nmap
  • Hydra
CATEGORY / 03DOMAIN

Defensive & Systems

Host-based defense, kernel-level networking, and system hardening.

  • iptables / Netfilter
  • Linux Kernel Internals
  • HIDS / IPS
  • Socket Programming
  • Linux Hardening
CATEGORY / 04DOMAIN

Cryptography

Post-quantum protocols, authenticated encryption, and key exchange.

  • NIST PQC (ML-KEM-768, ML-DSA-65)
  • X25519 ECDH
  • AES-256-GCM
  • Hybrid Key Derivation
CATEGORY / 05DOMAIN

Frameworks & ML

API development, mobile security apps, and ML pipelines.

  • FastAPI
  • Flask
  • Jetpack Compose
  • Keras / TensorFlow
  • Hugging Face Transformers
CATEGORY / 06DOMAIN

Platforms & Tools

Development environments, version control, and deployment.

  • Linux (Ubuntu, Kali, Debian)
  • Git / GitHub
  • Docker
  • Firebase Firestore
  • Android SDK

Education & Development

CURRENTSTUDIES

Cybersecurity Studies

Studying cybersecurity with coursework in cryptography, network security, operating systems, and software engineering. Active in CTF competitions focused on reverse engineering, network forensics, and web exploitation.

05 / CONTACT

Let's discuss security engineering.

I am open to internship opportunities, research collaborations, and conversations about cybersecurity engineering. Reach out through any of the channels below.