Hi, I'm Tara! I'm a CS student at Stanford interested in security, especially where the security we expect on paper breaks down in practice. Recently, I've been working on malware-model reliability, vulnerability intelligence, enterprise security, and security for autonomous AI systems.

GitHub | LinkedIn | Google Scholar | Email | Resume

Selected Results

Malware-model reliability

98.0% overall accuracy hid an 85.2% false-negative rate for one malware family.

Aggregate calibration looked excellent. Slice-level evaluation exposed a concentrated failure that the headline metrics missed.

Code Paper

Overall ECE

0.0031

Malicord ECE

0.602

CVE enrichment

A simple text model predicted eight security attributes from vulnerability descriptions with 86% average accuracy.

But its predicted score and severity contradicted each other on 1 in 5 test CVEs. Rare vulnerability types were also much harder to recover.

Code Paper

Government security guidance

Only 2 of 166 observed controls were universal across 10 deeply analyzed frameworks.

Even close security allies differed substantially in what they recommended enterprises prioritize.

See all research