Capra (original) (raw)

Capra is a research group at Cornell in the Computer Science and Electrical and Computer Engineering departments. Our research studies abstractions and efficiency through the interaction of programming languages and computer architecture.

Check out our ongoing research below or read news about the group. If you’re a Cornell undergraduate student, consider working with us!

Hardware Accelerator Generation

Filament, an HDL for Fearless Hardware Design

Filament is a new hardware design language that uses a substructural type system to reason about low-level programs and ensure that they generate correct and efficient hardware.

Calyx, an Infrastructure for Hardware Accelerator Compilers

We’re designing Calyx, an intermediate language (IL) and infrastructure for building compilers that generate hardware accelerators. Calyx works by representing both hardware-like structure and software-like control together. Calyx is a part of the LLVM CIRCT project and supports Cider and Pollen. You can try Calyx in your browser.

Dahlia, a Language for Predictable Accelerator Design

High-level synthesis (HLS) tools can translate C-like languages to hardware accelerators, but the semantic gap between software and hardware can yield unpredictable performance and semantics.Dahlia adds a substructural type system to model hardware resources and their constraints to statically reject HLS designs that make unpredictable area-latency trade-offs. You can try Dahlia in your browser.

Software-Defined Networking

Rio, a Language for Programmable Packet Scheduling

Prior work has explored programmable packet scheduling on internet routers. Here we are interested in exploring packet schedulers that can live on the FPGA portions of smartNICs, and decide how packets get scheduled to hosts that can consume those packets. We’re developing Rio, a domain-specific language for defining packet scheduling policies, and a compiler to translate policies down to FPGAs.

Graphics Programming

Gator: Geometry Types

We have identified a new category of geometry bugs that arise in graphics programming and other domains that have to deal with matrices and vectors. They arise when programmers lose track of the coordinate systems and reference frames that underpin the computation.Gator is a language for GPU shading with a type system that can eliminate geometry bugs and rule them out by generating correct-by-construction transformation code.

Braid, a Safe Heterogeneous Language for Real-Time Graphics

Braid is a programming language for heterogeneous programming, where a single source program targets different hardware units. We have applied it to real-time graphics programming on CPU–GPU systems. Braid compiles to WebGL, so you can try it out in your browser.

Search-Based Compilation for Digital Signal Processing

Digital signal processors (DSPs) are ubiquitous and energy efficient, but making them fast requires an expert programmer. The difficulty stems from their complex vector instruction sets and simple, in-order pipelines. To get the best results, programmers must carefully pack and move data in vector registers to enable compact execution. Diospyros uses equality saturation to automatically discover efficient vector packing schemes.

Vision/System Co-Design

Customizing JPEG Compression for Computer Vision

Image compression formats like JPEG are ubiquitous in computer vision, but they were designed for human perception—not for modern vision algorithms. We examine the potential for customizing JPEG compression for specific vision tasks, simultaneously improving compression the ratio and the accuracy.

Exploiting Temporal Redundancy for Live Computer Vision

Vision accelerators that run on real-time video process nearly identical frames at every time step. This project introduces activation motion compensation, a technique for approximately incremental acceleration of computer vision. It works by measuring motion in the input video and translating it to motion in the intermediate results of convolutional neural networks.

A Vision Mode for Efficient Image Capture

Most camera systems are optimized for photography, so they waste time and energy when they capture images for computer vision. This project designs a vision mode for cameras and their associated signal processing logic that saves energy by producing lower-quality, less-processed image data.

Archived Research

People

PhD & MS

Undergrad & MEng

Staff

Alumni

News

Mar 22, 2026

VDCores, Our new programming framework and runtime for async GPUs, is now opensource! Learn more in our technical blog post.

Dec 11, 2025

We are once again organizing LATTE, an ASPLOS workshop about PL and compilers for hardware design. Submit 2-page position papers by January 31.

Nov 25, 2025

“Parameterized Hardware Design with Latency-Abstract Interfaces,” our paper about safely composing hardware generators, will appear at ASPLOS 2026.

Sep 26, 2025

Adrian won a Bowers college teaching award.

Jun 27, 2025

What is Runtime Verification?

Jun 24, 2025

Rachit won SIGPLAN’s John C. Reynolds Doctoral Dissertation Award for 2025! Incredibly, he was also an Honorable Mention for the SIGARCH/IEEE CS TCCA Outstanding Dissertation Award.

May 13, 2025

Felicitations to Dr. Kevin Alarcón Negy on the occasion of his successful PhD defense! Kevin will soon start as a lecturer in the CS department at Princeton.

Apr 06, 2025

Congratulations to Dr. Rachit Nigam who graduated with his PhD and will start at MIT as a tenure-track professor leading the FLAME Lab!

Mar 21, 2025 by Griffin

Unfortunately you sometimes need to do the thing

Jan 09, 2025

We are again organizing LATTE, an ASPLOS workshop about PL and compilers for hardware design. Submit 2-page position papers by January 31.

Oct 20, 2024

Caleb and Pai’s paper on unifying static and dynamic approaches to accelerator compilation appeared at OOPSLA 2024.

Sep 26, 2024 by Ayaka

So I Taught a CS Course Over the Summer…

Aug 08, 2024 by Adrian

Geometry Bugs and Geometry Types

Aug 05, 2024

We’re excited that Kevin Laeufer is joining the lab as a research associate!

Aug 01, 2024

Congratulations to Drs. Dietrich Geisler and Neil Adit, both of whom graduated with their PhDs! Neil is starting a position at Meta, and Dietrich will be a faculty of instruction at Northwestern CS.

Jul 26, 2024 by Adrian

Bril: An Intermediate Language for Teaching Compilers

Jul 15, 2024 by Adrian

Automated Test-Case Reduction

May 28, 2024 by Adrian

One Weird Trick for Efficient Pangenomic Variation Graphs (and File Formats for Free)

May 25, 2024

Caleb won the 2024 Cornell Computer Science Prize for Academic Excellence!

May 15, 2024

Ayaka’s paper on a FIRRTL backend for Calyx appeared at OSDA 2024.

May 13, 2024 by Ayaka

2024 CRA-WP Grad Cohort for Women Report

Apr 30, 2024 by Adrian

Pangenomic Variation Graphs, and a Reference Data Model

Mar 15, 2024 by Griffin

So You've Decided to Start a Podcast (or you need to edit audio)

Feb 20, 2024 by Griffin

On Asking the Wrong Questions