Skip to main content

Mark Buckler

Applied Scientist


I’m currently an Applied Scientist at Amazon where I help this cute robot see the world with fast, accurate, and energy efficient computer vision. I recently graduated with a PhD in Electrical and Computer Engineering from Cornell University where I was advised by Prof Adrian Sampson.

My graduate research focused on efficient computer vision. I’ve found that abstractions can simplify the design process, but breaking down boundaries through hardware-software co-design can produce superior results. For this reason I see myself as a system creator rather than a hardware or software developer.

As an entrepreneurial engineer I quickly gravitated to both industrial and academic research. While finishing my M.S. at UMass Amherst I formed Firebrand Innovations as a way of monetizing intellectual property I developed while in high school. These days I can be found in Amazon’s robotics lab building powerful and exciting new computer vision systems.


  • Computer Vision
  • Machine Learning
  • Robotics
  • Computer Architecture
  • Embedded Systems


  • PhD in Electrical and Computer Engineering, 2019

    Cornell University

  • M.S. in Electrical and Computer Engineering, 2014

    University of Massachusetts, Amherst

  • B.S. in Electrical Engineering, 2012

    Rensselaer Polytechnic Institute

Selected Publications

  • , , , , EVA²: Exploiting Temporal Redundancy in Live Computer Vision, in International Symposium on Computer Architecture (ISCA).

    Details arXiv PDF Slides Video Project

  • , , , Reconfiguring the Imaging Pipeline for Computer Vision, in International Conference on Computer Vision (ICCV).

    Details DOI arXiv PDF Slides Code Project

  • , , , , Dynamic synchronizer flip-flop performance in FinFET technologies, in IEEE Symposium on Networks-on-Chip (NOCS).

    Details DOI PDF

  • , , Predictive synchronization for DVFS-enabled multi-processor systems, in IEEE Symposium on Quality Electronic Design (ISQED).

    Details DOI PDF

  • , , , Low-power networks-on-chip: Progress and remaining challenges, in IEEE Symposium on Low Power Electronics and Design (ISLPED).

    Details DOI PDF

Selected Patents

  • , , , Configurable image processing system and methods for operating a configurable image processing system for multiple applications, filed:

    Details Google Patents

  • , Continuous Frequency Measurement for Predictive Periodic Synchronization, filed: , granted:

    Details Google Patents USPTO

  • , , Predictive Periodic Synchronization Using Phase-Locked Loop Digital Ratio Updates, filed: , granted:

    Details Google Patents USPTO

  • , Synchronizer Circuits With Failure-Condition Detection and Correction, filed: , granted:

    Details Google Patents USPTO

  • , Video Conferencing, filed: , granted:

    Details Google Patents USPTO

Recent Posts

More Posts

This is the third in a series of three posts teaching you how to DJ! Part 1: What You Need to Start DJing Part 2: How to Learn DJ Basics Part 3: How to Make a DJ Mix Alright, you’ve got your gear and you’ve practiced a few transitions. Now is the time to dig into the meat of DJing: creating your mix. Below I describe my personal method, but as I’ve mentioned before there are many ways to succeed.


This is the second in a series of three posts teaching you how to DJ! Part 1: What You Need to Start DJing Part 2: How to Learn DJ Basics Part 3: How to Make a DJ Mix Great, you’ve got your musical knowledge, DJ software & hardware, and your music library. Lets get started on learning the fundamentals! First time setup Import your music library into your DJ software


Welcome! If you want to learn how to DJ, then you’re in the right place. This is the first in a series of three posts teaching you how to go from knowing nothing to being able to play at your first party. Below I share a few personal details about how I got started, but feel free to skip to “What you need to get started” if you just want technical details.


I’ve greatly enjoyed my time as a graduate research intern at DeepScale and so I was pleased when Forrest Iandola asked me to give a talk to his team on deep learning hardware. To keep the talk from being a boring list of slides recounting famous (but dry) neural network hardware papers, we decided to structure it like one of David Patterson’s talks which describe what not to do. The talk was well received, so I decided to reformat it into this blog post.


If you’ve read my post about patenting in academia and industry then you know why you might want to file a patent with your university or company. The first step to filing a patent in most organizations (after creating the invention of course!) is filling out an Invention Disclosure Form or IDF. These forms will be created, provided by, and reviewed by your organization’s technology transfer office or center for technology licensing.



Low Precision SqueezeDet

Simulated fixed point model evaluation for SqueezeDet

Approximate Vision Pipeline

Using a reversible imaging pipeline to optimize sensor and ISP design for computer vision

YouTube BoundingBox Dataset Downloader

Helpful scripts I wrote for downloading and parsing Google’s huge video dataset

Reversible Imaging Pipeline

A Halide implementation of a forward and reverse computational photography pipeline

Neural Network Accelerator with Logarithmic Number System

Hardware accelerator for neural network computation using the LNS

Configurable Imaging Sensor

Tapeout of a configurable and energy-proportional image sensor

Network-on-Chip Synchronization

My M.S. thesis on synchronization circuits and systems for multi-clock domain Networks-on-Chip


What started as a Science Fair project became Firebrand Innovation’s first product


ECE 3140 (Embedded Systems) Spring 2015 - Teaching Assistant - Cornell University


  • Seattle WA, USA