Alan Reiner, Machine Learning Developer in Columbia, MD, United States
Alan Reiner

Machine Learning Developer in Columbia, MD, United States

Member since January 17, 2020
Alan is a statistician, data scientist, and deep learning practitioner with over a decade of data science experience, including computer vision for missile defense, real-time cybersecurity attack detection analytics, and borrower creditworthiness classification. Alan is motivated by challenging problems, is highly adaptable, and is especially good at picking up new projects and making an immediate impact.
Alan is now available for hire

Portfolio

Experience

Location

Columbia, MD, United States

Availability

Part-time

Preferred Environment

TensorFlow, Scala, Python, Linux

The most amazing...

...project I've worked on was a fully-functional poker-bot, which was trained by watching 1,500 games of a professional poker player.

Employment

  • Natural Language Processing Engineer

    2020 - PRESENT
    Novetta Corp
    • Designed an NLP dialog system in Tensorflow 2.0 for non-technical users to create complex, structured queries. Achieved near-SOTA NLP performance on limited dataset (>95% F1-score on 6 NLU tasks).
    • Fine-tuned BERT with multi-task LSTM head models, multiple NLP input streams, self-attention. Substantial performance gains via augmentation, custom loss functions, CRFs, keras-tuner.
    • Coordinated dataset design, collection, cleaning and annotation with six annotators.
    • Developed an open-source solution for rapid, hotkey-based, multi-label annotations: https://github.com/etotheipi/keynotate.
    • Mentored team members on machine learning, TensorFlow, AI architectures, Git, and Docker.
    Technologies: Custom BERT, LSTM, Docker, Scikit-learn, Python 3, Keras, TensorFlow
  • Data Scientist | Software Engineer

    2016 - 2019
    IronNet Cybersecurity
    • Engineered crucial features to improve a DNS-tunneling detection algorithm which reduced false positives by 80% and increased computational efficiency by 70%.
    • Collected and analyzed data for detecting malicious meek (domain fronting) connections originating in customer networks.
    • Developed a domain-generation attack (DGA) detection algorithm using a variety of machine learning (ML) techniques, including LSTM for identifying randomly generated domains.
    • Served as the technical lead in the migration of the entire company product’s back end to Docker and Kubernetes (50+ microservices).
    • Developed scripts for full-stack rollouts in AWS, including TLS certifications, DNS routes, security groups, and so on; also integrated them into the CI/CD pipeline.
    Technologies: Amazon Web Services (AWS), Kubernetes, Docker, AWS, Apache Kafka, Machine Learning, Python, Spark, Scala
  • CEO | Lead Developer

    2013 - 2016
    Armory Technologies, Inc.
    • Developed an open-source Bitcoin wallet from scratch that innovated multiple usability and security features in the early days of Bitcoin.
    • It was a spare-time project for two years before receiving seed funding at a $4.2 million valuation in 2013.
    • Became a respected thought leader in the industry, and the software still protects over $2 billion (USD) worth of Bitcoin.
    • Managed five, full-time, remote developers.
    Technologies: Cryptography, NoSQL, PyQt, UX, UI, Bitcoin, C++, Python
  • Physicist, Computer Vision for Missile Defense

    2006 - 2013
    JHU Applied Physics Laboratory
    • Worked on algorithms for "Lethal Aimpoint," which uses an IR camera on the interceptor missile to detect the threat and identify where to hit it. Used a variety of image processing and statistical techniques to solve the problem (missile defense).
    • Developed an elaborate visualization tool that was used daily by dozens of engineers to examine and verify the results of missile simulations.
    • Created a CUDA/C++ algorithm to speed up image-processing tasks in our simulations by a factor of 50-200x. The simulations originally took three hours to run, reduced to a couple of minutes with this code (and installing GPUs in our clusters).
    • Developed statistical techniques to mitigate the effects of dead pixels in our IR cameras, and drive the requirements process for future camera/array production.
    Technologies: Video Processing, Image Processing, MATLAB, C++

Experience

  • Coursera ML/DL Courses
    https://www.coursera.org/specializations/aml

    I took 30 weeks of courses in the Advanced Machine Learning specialization on Coursera, in my spare time. This included Deep Learning, Bayesian Methods, Computer Vision, Reinforcement Learning and NLP.

    I worked on more than a dozen different ML/DL projects including object detection, face recognition, segmentation, image captioning, seq2seq models, text generation and learning to play Atari games with only raw screen pixels. I also gained direct experience with VAEs, GANs, U-Nets, transfer learning, LSTMs, Deep Q-Learning, A3C, and Bayesian methods such as EM and MCMC.

  • AI-powered Poker Bot

    I gained access to the text-file poker-hand histories of 1,500 online poker games played by a professional poker player. It used heavy feature engineering, statistics, neural networks, and SVMs to mimic that player's betting patterns for use in real-time play. The bot performed very well against other bots, though we were never able to deploy it in a real online poker site since the sites have gotten extremely good at detecting bots and confiscating funds.

  • LendingClub Peer-to-peer Investing

    LendingClub is a peer-to-peer platform to connect individual investors with borrowers who cannot get loans through a bank (or want a lower interest rate). The site has made public all the information of over one million historical loans, which makes it perfect for data mining and machine learning.

    I thoroughly analyzed and produced multiple different machine learning (ML) models to try to identify high-value borrowers and created an automated system to process new borrower applications and invest if they receive a high score.

  • The Many Applications of Gradient Descent in TensorFlow (Publication)
    TensorFlow is one of the leading tools for training deep learning models. Outside that space, it may seem intimidating and unnecessary, but it has many creative uses—like producing highly effective adversarial input for black-box AI systems.

Skills

  • Languages

    Python, C++, Scala, Python 3
  • Platforms

    Kubernetes, CUDA, Apache Kafka, Docker, Linux, Amazon Web Services (AWS)
  • Other

    Artificial Intelligence (AI), Computer Vision, Machine Learning, Natural Language Processing (NLP), Deep Learning, Data Engineering, Object Detection, Variational Autoencoders, Deep Neural Networks, Custom BERT, AWS, Image Processing, Video Processing, Bitcoin, UI, UX, Cryptography, Reinforcement Learning, Generative Adversarial Networks (GANs), LSTM Networks, Bayesian Inference & Modeling
  • Frameworks

    Spark
  • Libraries/APIs

    TensorFlow, Keras, Scikit-learn, LSTM, PyQt
  • Tools

    MATLAB
  • Industry Expertise

    Cybersecurity
  • Paradigms

    Functional Programming
  • Storage

    NoSQL

Education

  • Master's degree in Applied Mathematics—Statistics and Stochastic Processes
    2006 - 2009
    Johns Hopkins University - Baltimore, MD, USA
  • Bachelor's degree in Theoretical and Applied Mechanics (Engineering)
    2001 - 2006
    University of Illinois at Urbana-Champaign - Champaign, IL, USA
  • Bachelor's degree in Applied Mathematics
    2001 - 2006
    University of Illinois at Urbana-Champaign - Champaign, IL, USA

Certifications

  • Machine Learning and Deep Learning
    JANUARY 2019 - PRESENT
    National Research University of Higher School of Economics via Coursera

To view more profiles

Join Toptal
Share it with others