About Me
A record of who I am, what I build, the principles guiding my work, and my research trajectory in systems and browser security.
01 / ABOUT
Hey! I'm Devansh Mishra (often known as Tanishq online). I'm an engineer and systems builder focused on automations, web security, anti-fingerprinting primitives, and low-level scripting.
My engineering philosophy centers around demystifying systems: understanding software from the TCP/TLS packet handshake up to the JavaScript engine's bytecode interpreter. Rather than relying on superficial abstraction layers, I focus on how software behaves under adversarial conditions—how tracking networks profile users, how bot detection sensors measure client kinematics, and how robust software can be engineered to resist surveillance.
When not writing code or analyzing protocol telemetry, you will usually find me testing browser runtime internals, reading computer science papers, or scripting automated infrastructure.
02 / WHAT I'M BUILDING [Active Focus]
Website Fingerprint Spoofer v2.1
Adversarial browser privacy extension and proxy layer. Automatically introduces calibrated, imperceptible noise into Canvas 2D render buffers, masks WebGL GPU vendor strings, and perturbs Web Audio API oscillator synthesis—rendering device hashes non-deterministic across tracking sessions without breaking standard web apps.
StealthScrape Engine
High-concurrency web automation framework engineered to evade enterprise-grade anti-bot mitigation suites (Cloudflare Turnstile, DataDome, Akamai). Implements native V8 prototype patching, CDP artifact eradication, dynamic TLS JA3/JA4 profile rotation, and humanized biometric cursor movement simulation.
Autonomous Web Verification Lab
Deterministic testing platform that benchmarks browser stealth frameworks against production anti-abuse sensors, tracking entropy drift and telemetry flag triggers across Chromium updates.
03 / EDUCATION
Meerut Institute of Engineering & Technology (MIET)
2023 — Present04 / EXPERIENCE
Spearheading development of the open-source browser privacy project. Authored core mathematical perturbation algorithms for Canvas 2D and AudioContext, reviewed community contributions, and maintained documentation.
Reverse-engineered client-side sensor payloads across major bot mitigation providers. Developed automation pipelines emulating native human behavioral telemetry and protocol-level consistency.
Built asynchronous diagnostic tools, TLS fingerprint auditing utilities, and webhook queue engines in Python and TypeScript.
05 / RESEARCH [Focus Areas]
Investigating why binary API blocking breaks ~34% of standard web applications, and demonstrating that calibrated ±0.001% buffer dithering alters cryptographic hashes without perceptual human impact.
Analyzing AST-obfuscated sensor payloads to catalog client-side feature vectors including mouse curvature derivatives, focus ratios, and worker execution timing.
Designing hook patterns that reconstruct Function.prototype.toString descriptors and eliminate Chrome DevTools Protocol artifacts before initial evaluation.
06 / VALUES [Principles]
A portfolio should never try to perform tricks. If the systems, code, and ideas are genuinely solid, the interface should simply step aside and make them effortless to discover.
Defensive engineering should neutralize surveillance by default rather than asking users to sacrifice usability or break standard web experiences.
Don't just consume abstraction layers. Understand the C++ runtime underneath, the V8 bytecode execution, the TCP/TLS handshake packets, and the mathematical entropy vectors.
Working benchmarks, reproducible scripts, and open-source implementations always outweigh speculative claims.