96%
On curated virus corpus during evaluation
Virus Detection System
Prototyped a Windows-native malware detection engine featuring signature scanning, heuristic analysis, and quarantining for suspicious processes.
CourseworkPublic repositoryJan 2023 – Apr 2023IIT Jodhpur, India
Impact metrics
Key outcomes delivered for stakeholders
< 1%
After heuristic tuning and whitelist support
Highlights
Notable milestones and system improvements
- Reached 120k signature comparisons per second on benchmark dataset
- Implemented sandbox simulator that replayed malware behaviour for detection tuning
- Built exportable incident reports for security teams
Responsibilities
Where I created the most impact
- Developed multi-threaded scanning core leveraging memory-mapped files for signature throughput
- Hooked critical Win32 APIs using Microsoft Detours to observe file and registry operations in real time
- Authored heuristic scoring and quarantine workflow with CLI reporting dashboard
Project narrative
Snapshot of the project background, execution, and results
Project context
Needed a performant detection prototype to explore real-time malware interception without access to enterprise antiviruses.
Approach
Combined hash-based scanning with API hooking, enabling rapid signature checks alongside behavioural heuristics.
Impact
Demonstrated reliable detection in academic evaluations and provided a foundation for further research.
Stack
Tools and frameworks that powered the build
C/C++Microsoft DetoursNetworkSHA256