My Research
TL;DR: I'm interested in building compilers specifically to optimize Fully Homomorphic Encryption programs.
Fully Homomorphic Encryption refers to any number of encryption schemes that preserve the ring structure of their plaintext (i.e. if you encrypt two numbers and add or multiply the ciphertexts, you get back an encryption of the sum or product of the original numbers).
FHE is a powerful tool for for implementing arbitrary fixed-size arithmetic circuits over encrypted data, but the underlying computational model is very different from the one you're used to.
Taking these pecularities into account when compiling arithmetic FHE circuits is often challenging, and my research is interested in building compilers that automatically factor in all the FHE weirdness to produce efficient arithmetic circuits while letting the programmer concentrate on writing their programs.
At the moment, I'm specifically interested in developing vectorization techniques for FHE programs.
While normal FHE computation is very slow (like, several orders of magnitude slower than normal), it does provide the ability to vectorize, or pack multiple identical operations into a single step.
We have to be careful when vectorizing FHE programs, though, since at every step, all the data has to line up perfectly, and shuffling data around in FHE vectors is very expensive.
I'm currently working on building a compiler that automatically figures out both data placement and how to group together operations to convert a scalar arithmetic circuit into an efficient vector program.
Publications
Experience
Education
Internships
-
Microsoft Research (MSR)
Summer 2022
Seattle, WA
Worked on developing better cost models and scheduling optimizations for the MSCCL GPU communication collective compiler.
-
Amazon Robotics
Summer 2021
Austin, TX
Developed a DSL for expressing complex data validation rules, and created a bot that ran at code review time to flag data records that broke as a result of any change pushed to a validation model.
-
AWS Kinesis
Summer 2020
Palo Alto, CA
Designed and developed a feature to add support to Kinesis Firehose for streaming and batching Avro records directly into S3.
-
AWS Kinesis
Summer 2019
Palo Alto, CA
Onboarded Kinesis Firehose with AWS Service Quotas to provide a streamlined interface for customers to request usage limit increases. Deployed a serverless Lambda function to monitor and report usage Firehose statistics.
Talks
-
Coyote: A Compiler for Vectorizing Encrypted Arithmetic Circuits
6 October 2023
Midwestern PL Summit 2023
17 April 2023
Galois, Inc.
27 March 2023
ASPLOS 2023
-
Vectorized Secure Evaluation of Decision Forests
19 October 2022
Cornell University
21 October 2021
Splash! 2021
Contact
Elements
Text
This is bold and this is strong. This is italic and this is emphasized.
This is superscript text and this is subscript text.
This is underlined and this is code: for (;;) { ... }
. Finally, this is a link.
Heading Level 2
Heading Level 3
Heading Level 4
Heading Level 5
Heading Level 6
Blockquote
Fringilla nisl. Donec accumsan interdum nisi, quis tincidunt felis sagittis eget tempus euismod. Vestibulum ante ipsum primis in faucibus vestibulum. Blandit adipiscing eu felis iaculis volutpat ac adipiscing accumsan faucibus. Vestibulum ante ipsum primis in faucibus lorem ipsum dolor sit amet nullam adipiscing eu felis.
Preformatted
i = 0;
while (!deck.isInOrder()) {
print 'Iteration ' + i;
deck.shuffle();
i++;
}
print 'It took ' + i + ' iterations to sort the deck.';
Lists
Unordered
- Dolor pulvinar etiam.
- Sagittis adipiscing.
- Felis enim feugiat.
Alternate
- Dolor pulvinar etiam.
- Sagittis adipiscing.
- Felis enim feugiat.
Ordered
- Dolor pulvinar etiam.
- Etiam vel felis viverra.
- Felis enim feugiat.
- Dolor pulvinar etiam.
- Etiam vel felis lorem.
- Felis enim et feugiat.
Icons
Actions
Table
Default
Name |
Description |
Price |
Item One |
Ante turpis integer aliquet porttitor. |
29.99 |
Item Two |
Vis ac commodo adipiscing arcu aliquet. |
19.99 |
Item Three |
Morbi faucibus arcu accumsan lorem. |
29.99 |
Item Four |
Vitae integer tempus condimentum. |
19.99 |
Item Five |
Ante turpis integer aliquet porttitor. |
29.99 |
|
100.00 |
Alternate
Name |
Description |
Price |
Item One |
Ante turpis integer aliquet porttitor. |
29.99 |
Item Two |
Vis ac commodo adipiscing arcu aliquet. |
19.99 |
Item Three |
Morbi faucibus arcu accumsan lorem. |
29.99 |
Item Four |
Vitae integer tempus condimentum. |
19.99 |
Item Five |
Ante turpis integer aliquet porttitor. |
29.99 |
|
100.00 |