sidarthx0

Core developer on a real-time blockchain security monitoring system, building and maintaining production-grade security detectors across live EVM chains. Designed and implemented rule-based static & dynamic detectors like: Library tampering, Malicious mint/burn paths, transfer & approval blocking mechanisms, suspicious external calls & balance updates, etc. Improved detection coverage & reduced false positives using mix of rule-based & ML-driven approaches. Engineered opcode-, CFG-, and storage-level features for ML-based vulnerability detection models, and validated predictions against real-world exploits and false positives. Modeled real-world exploit flows like access control bypass, state manipulation attacks, input validation, etc. Built and extended code analysis pipelines (AST, CFG, bytecode parsing) and performed fuzzing and mainnet-fork simulations. Caught in production: traced OpenZeppelin library tampering to a $1M live scam on Base Network.

Worked primarily on Smart Contract Research, Internal Auditing & Development. Built On-Chain NFTs, marketplace, staking, renting and Raffle based NFT drops with Merkle tree authentication, Dutch Auction (GDAs and CGDA) Workflow POC’s. Conducted manual security reviews, identifying vulnerabilities and validating remediation strategies. Developed an automated Python CI/CD pipeline for NFT metadata, image stitching & rarity for MCL drop, which generated $1.2 million in launch revenue sustaining $100K – $150K monthly trading volumes.

Neural Machine Translation based Encoder-Decoder Attention Model to translate Japanese Documents. Building Deep learning models coordinating with Harness optimisation systems.

NLP Projects involve building of Language translation pipelines, Grammar Error Detection models. Computer Vision & Object Detection projects like digitalizing Aadhar cards, forms using Faster-RCNN model and OCR extractions.