Predict Lightning Strikes using Django and AWS
09-22, 15:10–15:40 (Europe/Lisbon), Auditorium

Can you predict lightning strikes using Django and AWS? Yes! Learn how to take an algorithm and idea from a Jupyter Notebook to production ready and cloud native.

Predicting a lightning strike with 99.6% accuracy requires advanced algorithms, expert developers and data — a lot of data. FLASH Scientific Technology pulls data from a variety of sources including weather radars and location-specific information like altitude, moisture, wind direction and temperature. It was imperative that FLASH develop an application that could pull, read and process the data both quickly and accurately.

In this case study and product demo, you’ll learn how to take an algorithm and idea from a Jupyter Notebook to production ready and cloud native.

  1. Optimize Python code structure
    1. As the Zen of Python states, “simple is better than complex.” The first part of this presentation will focus on the importance of reviewing and simplifying code to ensure faster data processing speeds and that the code is easy to both read and duplicate.
  2. Build fast, efficient and accurate APIs with Django
    1. Once the code is optimized, it’s ready for production. You’ll learn how we built an interface that could both request data from various web services and quickly make accurate predictions using FLASH’s proprietary and patent-issued algorithm. Specifically, this talk will feature:
      1. best practices and lessons learned, including insight into why we stored all of the Django application elements in a single library;
      2. why all of the data needed to be pulled directly from the cache; and
      3. why the algorithm was kept separate.
  3. Deploy serverless and scalable applications using AWS Lambda and AWS Fargate
    1. Finally, this talk will:
      1. showcase how the Django application we built integrates seamlessly with the serverless components of AWS including AWS Lambda and AWS Fargate;
      2. unveil the secrets of building Python Container Images for AWS Lambda (Hint: by using container images for Lambda code vs. manually submitting .zip files we saved significant time, money, and effort. We also reduced errors associated with manual changes.);
      3. review the importance of building applications to be serverless from the beginning.
  4. Product Demo
    1. Today, the application — which initially took minutes to make lightning strike predictions with 99.6% accuracy and a 15- to 25-minute lead time (saving time, money and lives) — can predict when and where lightning will strike in just seconds. The presentation will conclude with a demo of the product, which is deployed with GitLab CI/CD merge request pipeline.

Both beginner and intermediate developers will benefit from this talk, and it is ideal for developers wanting to learn how to use Django for complex predictive-technology projects.

Video: https://youtu.be/41S6bwdGHRo
Calvin Hendryx-Parker

Calvin Hendryx-Parker is the co-founder and CTO of Six Feet Up, a Python and Cloud expert consulting company that helps organizations build apps faster, innovate with AI/ML, simplify Big Data, and leverage Cloud technology. At Six Feet Up, Calvin establishes the company's technical vision and leads all aspects of the company's technology development. He provides the strategic vision for enhancing the offerings of the company and infrastructure, and he works with the team to set company priorities and implement processes that will help improve product and service development.

In 2019, Calvin was named an AWS Community Hero — one of only 24 Heroes in North America. He is the founder and host of the Python Web Conference, the largest worldwide event for web developers; the co-founder of IndyPy, the largest Python meetup in Indiana with 1,900+ members; and the founder of IndyAWS, the fastest growing cloud meetup in the state with 800+ members. Additionally, Calvin is the driving force behind LoudSwarm by Six Feet Up, a high impact virtual event platform that debuted in June 2020.

Outside of work, Calvin spends time tinkering with new devices like the AWS DeepRacer, CircuitPython and Raspberry Pi. He is an avid distance runner and ran the 2014 NYC Marathon to support the Innocence Project. Before the pandemic, Calvin and his family enjoyed annual extended trips to France where his wife Gabrielle, the CEO of Six Feet Up, is from. Calvin holds a Bachelor of Science from Purdue University.

Calvin is passionate about the open-source community, and he is regularly sought after to share his expertise — both at international conferences and in the media. Some of his recent contributions include:

  • PyCon (April 2022): Calvin presented “Bootstrapping a Local Python Environment”
  • Python Web Conference (March 2022): Calvin presented two talks including: “Predict Lightning Strikes using Django and AWS” and “Data Pipeline Modernization at Scale”
  • Datanami (October 2021): Calvin was featured as an expert in the article, “What’s Driving Python’s Massive Popularity?”
  • DjangoCon (October 2021): Calvin presented on “Bootstrapping your Local Python Environment.”
  • PyOhio (August 2021): Calvin presented on how to hack Django Channels to seamlessly integrate Slack into the virtual event platform via “Hacking Django Channels for Fun (and Profit).”
  • DjangoCon Europe (June 2021): Calvin presented on how to use containers to implement a process that focuses on the developer experience and application delivery via “Deploying a Django Virtual Event Platform Using Containers and Terraform.”
  • App Developer Magazine (May 2021): Calvin shares why Python — with its “batteries-included” standard library — is one of the best languages for both beginner and advanced developers in the article, “4 Reasons Python is Taking Over the World.”
  • Cloud Field Day 10 (April 2021): Calvin was selected as a Field Day Delegate and invited to participate in Cloud Field Day 10 — a multi-day discussion with leaders and product managers from enterprise IT companies including NetApp, Dell Technologies, VMware, Oracle, Scality, Veeam, Intel, Komprise and StorageOS.