Recently graduated engineer building developer tools, resilient systems, and security focused infrastructure.

github

Brahadeesh V  • Peerlist

<h3>About ThreatLib</h3>ThreatLib is a platform-agnostic account risk scoring engine and SDK designed for backend services that require consistent, auditable, and privacy-preserving abuse detection across products. It accepts heterogeneous account, device, network, behavior, graph, payment, and content metadata. The engine normalizes events through platform adapters, runs detector evidence through a dependency-aware orchestration graph, and returns a risk score, confidence band, action recommendation, and feature-level restriction map. ThreatLib is built to handle environments where missing data is common, ensuring that the absence of evidence does not equate to evidence of legitimacy, which is central to its safety model.<h3>Key Features:</h3><ul><li>Platform Agnostic: Works across various platforms by using adapters for normalization.</li><li>Privacy-First Risk Assessments: Computes risk scores while adhering to strict PII isolation.</li><li>Dempster-Shafer Theory: Fuses independent behavioral, device, and network signals for robust scoring.</li><li>Cold Start Mitigation: Provides effective scoring even with limited historical data.</li><li>Configurable Action Tiers: Allows for flexible response strategies based on risk levels.</li><li>Shadow Mode: Enables risk assessment computation and review without immediate enforcement.</li><li>Extensible Architecture: Supports integration of developer-supplied ML model plugins.</li><li>Comprehensive API and SDK: Offers interfaces for scoring, event ingestion, reporting, and more.</li><li>Advanced Analytics: Provides detailed metrics for model performance, detector health, and replay simulations.</li><li>Domain-Specific Modes: Includes tailored configurations for social media, chat apps, and gaming.</li></ul>ThreatLib is built with a layered system architecture, including contracts, platform adapters, independent and interdependent detectors, risk synthesis, an action engine, and various interfaces. It prioritizes core invariants such as returning uncertainty for missing input, logging all scoring events, and ensuring data privacy by not persisting plaintext PII.

Threatlib

<h3>ContractGuard</h3>ContractGuard is a VS Code extension for finding security and reliability issues in code, configuration files, queries, Dockerfiles, and dependency files directly inside the editor. It focuses on practical analysis during development without requiring external services or separate scanning workflows.The extension supports scanning both the current file and the full workspace. Findings are shown as inline diagnostics in the editor and inside a dedicated explorer view so issues can be reviewed and navigated efficiently. ContractGuard also preserves existing workspace findings during incremental rescans of saved files to avoid losing previous results.ContractGuard checks for a wide range of problems across source code and project configuration files, including JSON schema drift, SQL query risks, regex patterns that may lead to ReDoS issues, hardcoded secrets, PII exposure, insecure configuration settings, Dockerfile security problems, and dependency vulnerabilities in Python and Node.js projects.The extension includes SARIF and JSON export support for external tooling and CI workflows. It also provides scan details and runtime information through the VS Code output channel along with an overall security score shown in the status bar.<h3>Features</h3><ul><li>Scan the current file</li><li>Scan the full workspace</li><li>Show findings in a dedicated explorer view</li><li>Publish inline diagnostics in the editor</li><li>Export findings as SARIF</li><li>Export findings as JSON</li><li>Disable selected rules from diagnostics or findings</li><li>Preserve workspace findings during incremental rescans</li><li>Show scan and runtime details in the output channel</li><li>Display an overall security score in the status bar</li><li>Filter low confidence fixture, documentation, and test findings by default</li><li>Continue scans even if one analyzer encounters a runtime issue</li></ul><h3>Checks Performed</h3><ul><li>JSON schema drift including type mismatches, nullable inconsistencies, and optional field differences</li><li>SQL query risks such as unsafe patterns and destructive queries without safeguards</li><li>Regex complexity and potential ReDoS risks</li><li>Hardcoded secrets including API keys, tokens, private keys, database URLs, and JWTs</li><li>PII exposure including emails, phone numbers, credit card patterns, SSNs, and dates of birth</li><li>Insecure configuration such as debug enabled deployments, weak defaults, open CORS policies, and disabled TLS</li><li>Dockerfile issues including root execution, floating image tags, embedded secrets, and SSH exposure</li><li>Dependency vulnerabilities in Python and Node.js dependency files</li></ul><h3>Commands</h3><ul><li>ContractGuard: Scan Workspace</li><li>ContractGuard: Scan Current File</li><li>ContractGuard: Export SARIF</li><li>ContractGuard: Export Findings JSON</li><li>ContractGuard: Clear Findings</li><li>ContractGuard: Install Python Runtime Dependencies</li><li>ContractGuard: Show Output</li></ul><h3>Extension Settings</h3><ul><li>contractguard.pythonPath</li><li>contractguard.scanOnSave</li><li>contractguard.scanDebounceMs</li><li>contractguard.scanOnSaveScope</li><li>contractguard.scanTimeoutMs</li><li>contractguard.enabledAnalyzers</li><li>contractguard.disabledRules</li><li>contractguard.minimumSeverity</li><li>contractguard.minimumConfidence</li><li>contractguard.includeFixtures</li><li>contractguard.rulesDirectory</li><li>contractguard.sqlExplainDatabase</li></ul><h3>Notes</h3><ul><li>Analysis runs locally</li><li>Default minimum confidence is set to medium</li><li>Use low confidence mode for broader audit style scans</li><li>Fixture, documentation, and test findings can be included through settings</li><li>SARIF export can be used with CI and external security tooling</li></ul>

Contract-Guard

A simple custom crafted PCB circuit to monitor various bags that contain fluids in hospitals

Bag monitoring system

This project was built following the seemingly non stop train accidents in India which occurred seemingly due to faulty signaling and track switches. This project aimed to provide error free signaling, taking various inputs like presence of trains in tracks, current position of the track switch and the priority that needs to be given to each train. This project brought my digital circuit design skills and reasoning abilities to test. I had to work day and night to bring down the signals that could allow movement of trains in each of the 128 possible cases with the input variables for a place with four tracks which were connected by track switches. I then redesigned the system to reduce complexity and deleted duplicates in the new design which brought the number of possible cases to 32. This was followed by solving K-Maps and Equations to find the most efficient expression for each output variable with minimum input term optimization, The designed circuit was realized and tested followed by the appreciation for the best project in the class during the semester project expo for third semester.

Collision Free Railway Signaling

Involved in the design and verification of a Instrumentation amplifier circuit for ECG signals

ECG recording setup

Bringing fields to the forefront of technology

FarmCare

Your fitness companion https://brahadeeshunstop.wixsite.com/fun---fit

Fun&FIt

This was the first set of projects that I had undertaken with my friends. We completed 3 projects together in the second semester to practically put to use, the knowledge gained by learning the subject, electron devices. The projects were CE Audio Amplifier, Clap Switch, Automatic Light Using Light Dependent Resistor

Mini project

Involved in the design, simulation and development of a novel single axis MEMS accelerometer fabricated using nickel with reduced cross sensitivity

Nickel Based MEMS Aceelerometer

An innovative and inclusive approach to disaster management

SafeReach

Developed a solar-powered, motion-activated street light with auto-orienting panels for maximum efficiency. Integrated sensors for air quality, weather, fire, and traffic monitoring, aiding smart city planning and emergency response. Added an SOS system for enhanced safety and disaster support.

Smart Street Light System

SmartVolt

A comaparitive study of various ML algorithms to predict Solar generation output