CUT TO:
INT. PROJECT ARCHIVE — STORYBOARD ROOM
The USER opens Mini-LOP.
DIVAKAR DESSAI
CUT TO:
The USER opens Mini-LOP.
DIVAKAR DESSAI
CASE FILE / COMP3142 / Security / Software Testing
Built a small AFL-inspired coverage-guided fuzzer that mutates program inputs and retains test cases that discover new execution behaviour.
01–02 / OPENING SEQUENCE
01 / Establishing Shot
Manually selected test inputs can miss unexpected execution paths, particularly in software with large or unusual input spaces.
02 / Wide Shot
The project explored coverage-guided fuzzing inspired by AFL-style security-testing systems.
03 / CHARACTER NOTE
Subject
Divakar Dessai
Production
Mini-LOP
Take
03 / Role
Role notes
DIVAKAR DESSAI
04 / CLOSE-UP
Generating random inputs alone is inefficient; the system needed a mechanism for recognising which mutations were worth exploring further.
05 / TRACKING SHOT
A plan emerges.
The fuzzer repeatedly selected corpus inputs, mutated them, executed the target and used coverage information to determine whether the generated input should be retained.
06 / INSERT SHOTS
The system takes shape.
07 / DIRECTOR'S NOTES
Things we decided along the way
01
Used coverage feedback to guide exploration.
02
Retained only mutations that contributed useful new behaviour.
03
Repeated mutation of promising inputs to progressively expand coverage.
design decisions
somewhere mid-build
08 / RETAKES
Naturally, not everything cooperates.
09 / FINAL SHOT
Built a small coverage-guided fuzzing system.
Gained practical exposure to automated security testing.
Learned how feedback can transform random mutation into directed program exploration.
10 / PRODUCTION NOTES
The tools behind the scenes.
11 / BEHIND THE SCENES
FADE OUT.
USER closes the file.
One project down. A few more stories left.
DIVAKAR DESSAI