@sunnychokshi

Sunny Chokshi

@sunnychokshiMassachusetts, USAJoined October 2026

Doctoral student researching LLM security

About

I am a doctoral student at the University of the Cumberlands. My dissertation looks at the adversarial robustness of large language models in cybersecurity applications. I write short notes on experiments I ran myself, mostly LLM code review and crypto API misuse, with the numbers and a link to the preprint or dataset behind each one.

Available for

Nothing here yet.

Sunny Chokshi's blogs

Sunny Chokshi: LLM Security Notessunnychokshi.hashnode.dev2 posts

Detection rate alone ranks the worst LLM code reviewers first

When I tested 16 open LLM code reviewers, the first table in my paper showed which of them separate secure code from vulnerable code at all. It is worth reading on its own. The setup I used 180 Python

2h ago · 4 min readsunnychokshi.hashnode.dev

A hex literal made LLM code reviewers raise false alarms

I wanted to know whether an LLM code reviewer follows what the code does or what the code looks like. Cryptographic API misuse is a clean place to ask, because the same insecure choice can be written

2h ago · 3 min readsunnychokshi.hashnode.dev

Sunny Chokshi (@sunnychokshi) | Hashnode

Sunny Chokshi

About

Available for

Sunny Chokshi's blogs

Detection rate alone ranks the worst LLM code reviewers first

A hex literal made LLM code reviewers raise false alarms

Search Hashnode

Sunny Chokshi

About

Available for

Sunny Chokshi's blogs

Detection rate alone ranks the worst LLM code reviewers first

A hex literal made LLM code reviewers raise false alarms