Liam Connell, Software Developer in Chicago, IL, United States
Liam Connell

Software Developer in Chicago, IL, United States

Member since November 16, 2018
Liam has 4 years experience in all things data, including data engineering roles, machine/deep learning projects, and the harnessing of the cloud computing tools. With those three pillars of data, he develops powerful tools that can be applied to either business processes, strategic decision making, or consumer products. Liam prides himself on strong two-way communication, alignment with client goals, and the highest standards.
Liam is now available for hire

Portfolio

  • Auth0
    Python, Redshift, Airflow, AWS, Terraform

Experience

  • Python, 4 years

Location

Chicago, IL, United States

Availability

Part-time

Preferred Environment

OS X, git, Python virtualenv, Terraform AWS

The most amazing...

...thing I've coded is a variational auto-encoder that uses encoding gradients to generate synthetic data and computationally approximate the Wasserstein distance.

Employment

  • Data Engineer

    2016 - PRESENT
    Auth0
    • Created an ML-based customer health score that was able to improve sales and customer success personnel allocation by 80%.
    • Designed and developed an over-quota tracking process that drove a campaign to half a million dollars in increased ARR, 10% of the quarter’s ARR gains.
    • Developed and maintained a Redshift data warehouse along the entire pipeline: Coordinating with other engineering teams for designing import processes; ETL design and implementation; Dimensional modeling; Coordinating with the business community to ensure effective use of the data warehouse; designing Machine Learning processes to aid in decision making and gain insights.
    Technologies: Python, Redshift, Airflow, AWS, Terraform

Experience

  • A Tour Through TensorFlow with Financial Data (Development)
    https://liamconnell.github.io/jekyll/update/2016/07/18/a-tour-through-tensorflow-with-financial-data.html

    A tutorial that runs through implementations of Deep Learning algorithms as applied to the problem of stock market prediction using TensorFlow. It takes the philosophy of using TensorFlow's lowest level tools in order to build a solid understanding of auto-gradient software and the ML algorithms themselves.

    This project has been very popular since it was released, especially since at the time of release (early 2016), TensorFlow was a new technology and tutorials had stayed comfortably in the sphere of toy implementations like MNIST digit recognition. I regularly responded to monthly emails from researchers attempting to replicate the code, many of whom were Ph.D. candidates themselves.

Skills

  • Languages

    Python, SQL, R
  • Libraries/APIs

    TensorFlow, NumPy, Scikit-learn, Keras, Pandas, PySpark, Spark ML
  • Storage

    PostgreSQL, Redshift, MongoDB
  • Paradigms

    ETL
  • Other

    Machine Learning, Teaching, Data Warehousing, Online Tutoring, Deep Learning, Artificial Intelligence (AI), Generative Adversarial Networks (GANs), Variational Autoencoders, Image Recognition, AWS Cloud Architecture, Natural Language Processing (NLP), Predictive Modeling, Financial Modeling, Reinforcement Learning, Deep Reinforcement Learning
  • Frameworks

    Django, Django REST Framework
  • Tools

    Terraform, Apache Airflow

Education

  • Bachelor's degree in Mathematics
    2011 - 2015
    Colby College - Waterville, Maine

To view more profiles

Join Toptal
I really like this profile
Share it with others