David Eigen - Curriculum Vitae

Email: de@deigen.net


Google Scholar page

Resume as PDF

Publication Highlights

(full list below)
Progressive Checkerboards for Autoregressive Multiscale Image Generation
David Eigen
arXiv preprint 2026 (pdf) (github)
Finding Task-Relevant Features for Few-Shot Learning by Category Traversal
Hongyang Li, David Eigen, Samuel Dodge, Matthew Zeiler and Xiaogang Wang
CVPR 2019 (pdf) (github)
Gradient Agreement as an Optimization Objective for Meta-Learning
Amir Erfan Eshratifar, David Eigen and Massoud Pedram
NeurIPS Meta-Learning Workshop 2018 (pdf)
Predicting Depth, Surface Normals and Semantic Labels with a Common Multi-Scale Convolutional Architecture
David Eigen and Rob Fergus
ICCV 2015 (pdf)
Depth Map Prediction from a Single Image using a Multi-Scale Deep Network
David Eigen, Christian Puhrsch and Rob Fergus
NIPS 2014 (pdf)
OverFeat: Integrated Recognition, Localization and Detection using Convolutional Networks
Pierre Sermanet, David Eigen, Xiang Zhang, Michael Mathieu, Rob Fergus and Yann LeCun
ICLR 2014 (pdf)
Learning Factored Representations in a Deep Mixture of Experts
David Eigen, Marc'Aurelio Ranzato and Ilya Sutskever
ICLR Workshops 2014 (pdf)
Restoring An Image Taken Through a Window Covered with Dirt or Rain
David Eigen, Dilip Krishnan and Rob Fergus
ICCV 2013 (pdf)

Project Highlights

Multi-Cloud Model Inference Deployment & Scaling
at Clarifai, 2020-25

System to deploy and scale models on demand. Handles unexpected bursts and slowly adapting traffic, efficiently scales and shares GPUs between models. Streaming protocols allow workers to run on any network, enabling both customer on-site models and use of lowest-cost hardware providers.

Object Detection for Aerial Video
at Clarifai, 2017-20

Object detector for realtime detection on aerial videos. Led ML team and developed improvements to detection methods, data cleaning, measurement and iteration. Top-performing system in federal contracts over multiple years.

Interactive Image Model Training
at Clarifai, 2015-17

Developed fast classifier training used by customers and internal teams to quickly build many classifiers in diverse applications. Includes embeddings, quantization, data balancing, and network architecture. Benchmarked and improved based on customers' uses in deployments for improvements.

Image Content Moderation using Weak Labels
at Clarifai, 2015

Created industry-leading classifier for image-based content moderation and filtering, using target labels created automatically from a word-based classifier applied to weak labels and user-supplied text.

Automated Web Content Classifier
at IronPort/Cisco, 2009

Created a system to automatically classify web page content into categories.

NVLog Filesystem Journal
at NetApp, 2007

Rewrote filesystem journal to eliminate lock contention, leading to over 10% higher total system throughput.

Visualization for Differential Geometry
as RA with Prof. Banchoff, Brown Univ., 2000-2004

Created a software package for creating interactive differential geometry visualizations, and produced class labs and demonstrations using this software.

Paper Reviewer
2013 - present

Official submission reviewer for NeurIPS, ICLR, TMLR, annually; other conferences and journals on occasion.

Work and Research Experience

Clarifai
Principal Scientist, 2020 - 2025
Research Scientist, 2015 - 2020
Google Brain
Research Intern, 2013
New York University
Computer Science Dept, Courant Institute
PhD Student, 2010 - 2015
Cisco IronPort Systems
Software Engineer, 2007 - 2010
NetApp
Software Engineer, 2005 - 2007
Brown University
Research Assistant
with Prof. Thomas Banchoff, Mathematics Dept.
2000 - 2003

Education

New York University, New York, NY
Computer Science Dept, Courant Institute
Ph.D. 2015
Thesis: Predicting Images using Convolutional Networks: Visual Scene Understanding with Pixel Maps
Brown University, Providence, RI
Sc.M. Computer Science, 2004
Brown University, Providence, RI
Sc.B. Mathematics - Computer Science, 2003

Additional Projects

Compute Orchestration Protocols
at Clarifai, 2024-25

Advised in prototyping, design and development of "Compute Orchestration" protocols for model serving, including multi-cloud/multi-cluster workers, load balancing across workers, and scaling within clusters, along with simple RPC interface based on function introspection. Supports streaming as well as unary RPCs.

Video Inference Pipeline Engine
at Clarifai, 2025

Simple and fast video pipeline engine in python, supporting real-time and faster multi-stage model inference applications (e.g. detect and track). Automatic rate adjustment, parallel execution and pipelining in small, easily importable python package under 1K lines. (github)

Model Deployment Encapsulation
at Clarifai, 2023-24

Designed and led development of model packaging and encapsulation, including input/output formats, dependency installs and runtime resource estimates. Models can be shipped between clusters, including air-gapped environments, without being tied to container base images.

Model Deployment Scaling
at Clarifai, 2020-23

Developed system to automatically scale model deployments in a cloud environment based on inference request load. Can load and scale up neural network models from zero on demand to handle both unexpected bursts and slowly adapting traffic, and efficiently share GPUs between models.

LLM Inference and Training Integration
at Clarifai, 2023

Led efforts to integrate third-party open source LLMs into our training and inference pipelines, including larger GPU deployments, profiling and LoRA adapter training.

Model Training API
at Clarifai, 2018-23

Led efforts to integrate our training and experimentation system (see below) with our data platform to automatically create image-based object detection and classification models in multiple cluster environments. Handles data validation, model training, evaluation, and inference deployment. Defined benchmarks to measure accuracy and speed for different types of models, and found best price/performance points.

Estimating Camera FOV, Pitch and Roll from a Single Image
at Clarifai, 2021

Developed prototype method to estimate camera field of view, pitch and roll from single image. Located and assembled initial datasets and models, simplified and improved models adding normalized angle regression and NLL loss. Combined FOV with horizon estimates to find pitch with its estimated certainty.

Video Object Detection Streaming Engine
at Clarifai, 2020-21

Wrote core execution engine for video object detection and tracking: frame buffering, parallel inference, serialized tracking and completion, and interface with custom video streaming protocols used by customer.

Few-Shot Learning Research Projects
at Clarifai, 2017-19

Mentored interns on projects in few-shot learning. Published works at CVPR 2019 and NeurIPS Meta-Learning Workshop 2018.

Object Detection Neural Network and Code Framework
at Clarifai, 2016-18

Wrote object detection code framework for use with in-house neural network library and tensorflow. Created object detection models performing at state-of-the-art accuracy and ~1.5x faster compared to concurrently developed opensource object detectors.

Experiment and Training Infrastructure
at Clarifai, 2016-18

Built job scheduler and experiment tracking system targeted for ML model building, comparison, reproducibility and change tracking. Allowed team members to independently run modified codebases in our cluster, encapsulating most dependencies and automatically tracking changes for reproducibility.

Logo Recognition from Synthetic Data
at Clarifai, 2016

Built a system to train detection models to recognize logos in images based on synthetic data; filed patent application.

Face Detection and Recognition
at Clarifai, 2016

Created industry-competitive face detection and recognition system, based on data collected from publicly available sources. Initial labeling for the detector based on combinations of open-source detectors with data filtering, and refined with hard example mining.

Image Content Moderation using Weak Labels
at Clarifai, 2015

Created industry-leading classifier for image-based content moderation and filtering, using target labels created automatically from a word-based classifier applied to weak labels and user-supplied text.

Sender IP Reputation from Spam Trap Rates
at IronPort/Cisco, 2010

Created a system to classify IP addresses as likely spam or ham senders for email based on recent trap rates, using as input live streams of spam trap hits and overall mail volume estimates.

Web Reputation, Telemetry and Corpus
at IronPort/Cisco, 2007 - 2009

Datasource aggregator to score HTTP requests according to the chance of malicious content. Automatically fed back traffic samples from deployments to improve accuracy and coverage.

Updateable Web Reputation Client
at IronPort/Cisco, 2008

Extracted static client code and dependencies into a dynamically updateable package, taking into account potential differences in system libraries and hardware architecture between client platform releases. Distinct from system upgrades, the engine is automatically updated live on the appliance to the current provisioned version, without any input or interaction by the administrator.

Dynamic Microcores
at NetApp, 2007

Outlined a design for dynamic microcores, a reporting and debugging feature. Microcores are partial coredump files, only a few megabytes. The project aimed to let engineers write descriptive recipes to identify memory regions, and trigger microcore generation upon hitting system events. For example, if a system message warns about possible corruption in a block, one could identify interesting memory regions relative to the in-core structures for the block and inode in the message.

Effects of Interaction on Human Memory
Master's Thesis, 2004
at Brown Univ., 2004
(pdf)

Investigated the question of whether the use of different interaction techniques might impact the memory of a user. I designed a set of two experiments to address this: The first, a preliminary study confirming a well-known difference in performance between positional- and velocity-based controls, helped to verify my experimental methodology. The second, a comparison between three interaction modes in an immersive environment, was statistically inconclusive. Anecdotal evidence, however, suggested that for many subjects, memory performance improved with a full-body walking interaction.

Visualizing Deep Brain Stimulation Settings in Obsessive Compulsive Disorder
with Daniel Grollman, David Laidlaw, Benjamin Greenberg, Erin Einbinder
SIGGRAPH Poster 2004
at Brown Univ., 2003-2004
(www) (abstract pdf)

Wrote and reviewed proposals for a 6-week project in a mock grant proposal process during a class on scientific visualization. Carried out research on my project on electrode parameter settings in deep brain stimulation for obsessive compulsive disorder, collaborating with Benjamin Greenberg, a psychiatrist at Butler Hospital. Presented a poster of this project in SIGGRAPH 2004.

Visualization for Differential Geometry
Senior Thesis, 2003
work done as Research Assistant with Prof. Banchoff, Brown Univ., 2000-2003
(www) (pdf)

Created a software package for creating interactive differential geometry visualizations, and produced class labs and demonstrations using this software. Staff and students continue to use this package in new applications and to explore mathematical concepts in several courses at Brown, including differential geometry, combinatorial topology, calculus, geometry, and linear algebra. It has also been used by Prof. Banchoff in classes at UCLA, Notre Dame, University of Georgia, and Stanford.

Older Projects and Class Projects

Patents

System and method for facilitating graphic-recognition training of a recognition model
David Eigen, Matthew Zeiler
US Patent 11417130 (filed 2020; granted 2022)
Prediction-model-based mapping and/or search using a multi-data-type vector space
Matthew Zeiler, David Eigen, Ryan Compton, Christopher Fox
US Patent 11281962 (filed 2017; granted 2022)
System and method for facilitating logo-recognition training of a recognition model
David Eigen, Matthew Zeiler
US Patents 10163043, 10776675 (filed 2017; granted 2018, 2020)
System, method and computer-accessible medium for restoring an image taken through a window
Rob Fergus, David Eigen, Dilip Krishnan
US Patents 9373160, 9672601 (filed 2014; granted 2016, 2017)
Method and Apparatus for Generating Dynamic Microcores
David Eigen, David Grunwald
US Patent 7783932 (filed 2007; granted 2010)

All Publications

Progressive Checkerboards for Autoregressive Multiscale Image Generation
David Eigen
arXiv preprint 2026 (pdf) (github)
Enhancing Worldwide Image Geolocation by Ensembling Satellite-Based Ground-Level Attribute Predictors
Michael J Bianco, David Eigen, Michael Gormish
arXiv preprint 2024 (pdf)
Deep learning in medical image analysis: introduction to underlying principles and reviewer guide using diagnostic case studies in paediatrics
Constance Dubois, David Eigen, Emmanuel Delmas, Margot Einfalt, Clara Lemaçon, Laureline Berteloot, Patrick M Bossuyt, David Drummond, Pauline Scherdel, François Simon, Héloïse Torchin, Yasaman Vali, Isabelle Bloch and Jérémie F Cohen
bmj 2024 (pdf)
Development and validation of a smartphone-based deep-learning-enabled system to detect middle-ear conditions in otoscopic images
Constance Dubois, David Eigen, François Simon, Vincent Couloigner, Michael Gormish, Martin Chalumeau, Laurent Schmoll and Jérémie F. Cohen
npj Digital Medicine 2024 (pdf)
Efficient Training of Deep Convolutional Neural Networks by Augmentation in Embedding Space
Mohammad Saeed Abrishami, Amir Erfan Eshratifar, David Eigen, Yanzhi Wang, Shahin Nazarian, Massoud Pedram
arXiv preprint 2020 (pdf)
Finding Task-Relevant Features for Few-Shot Learning by Category Traversal
Hongyang Li, David Eigen, Samuel Dodge, Matthew Zeiler and Xiaogang Wang
CVPR 2019 (pdf) (github)
A Meta-Learning Approach for Custom Model Training
Amir Erfan Eshratifar, Mohammad Saeed Abrishami, David Eigen and Massoud Pedram
AAAI Student Abstract Track 2019 (pdf)
Gradient Agreement as an Optimization Objective for Meta-Learning
Amir Erfan Eshratifar, David Eigen and Massoud Pedram
NeurIPS Meta-Learning Workshop 2018 (pdf)
Predicting Depth, Surface Normals and Semantic Labels with a Common Multi-Scale Convolutional Architecture
David Eigen and Rob Fergus
ICCV 2015 (pdf)
Predicting Images using Convolutional Networks: Visual Scene Understanding with Pixel Maps
PhD Thesis, 2015 (pdf)
Unsupervised Learning of Spatiotemporally Coherent Metrics
Ross Goroshin, Joan Bruna, Jonathan Tompson, David Eigen and Yann LeCun
ICCV 2015 (pdf)
End-to-End Integration of a Convolutional Network, Deformable Parts Model and Non-Maximum Suppression
Li Wan, David Eigen and Rob Fergus
CVPR 2015 (pdf)
Depth Map Prediction from a Single Image using a Multi-Scale Deep Network
David Eigen, Christian Puhrsch and Rob Fergus
NIPS 2014 (pdf)
OverFeat: Integrated Recognition, Localization and Detection using Convolutional Networks
Pierre Sermanet, David Eigen, Xiang Zhang, Michael Mathieu, Rob Fergus and Yann LeCun
ICLR 2014 (pdf)
Learning Factored Representations in a Deep Mixture of Experts
David Eigen, Marc'Aurelio Ranzato and Ilya Sutskever
ICLR Workshops 2014 (pdf)
Understanding Deep Architectures using a Recursive Convolutional Network
David Eigen, Jason Rolfe, Rob Fergus and Yann LeCun
ICLR Workshops 2014 (pdf)
Restoring An Image Taken Through a Window Covered with Dirt or Rain
David Eigen, Dilip Krishnan and Rob Fergus
ICCV 2013 (pdf)
Nonparametric Image Parsing using Adaptive Neighbor Sets
David Eigen and Rob Fergus
CVPR 2012 (pdf)
Visualizing Deep Brain Stimulation Settings in Obsessive Compulsive Disorder
David Eigen, Daniel Grollman, David Laidlaw, Benjamin Greenberg, Erin Einbinder
SIGGRAPH Poster 2004 (www) (abstract pdf)
Effects of Interaction on Human Memory
with David Laidlaw
Master's Project, 2004 (pdf)
Visualization for Differential Geometry
with Thomas Banchoff
Senior Thesis, 2003 (www) (pdf)