Introducing Nodifly: Real-Time Coding Notification API

Stephen Hsu
Jul 11 · 5 min read

Data science is the “hottest job of the 21st Century”… but like anything that’s a dream, there comes the maintenance and unexpected headaches soon after. What’s often overlooked of the hottest job is the ugly, tedious side — waiting for queries to get the data needed for models, waiting for seemingly endless epoch training, and then more waiting just to see the model timed out or had an undefined variable. Nodifly was created out of this pain and is the solution to all of those problems - as the first simple to use real-time notification system designed by programmers for programmers.

In this post, we’ll be explaining the following:

  1. What does Nodifly do and how does it solve those problems?
  2. How do I use it?
  3. Where and how do I sign up?
  4. What’s in the future for Nodifly?
  5. What does Nodifly do and how does it solve those problems?

“Don’t even ask how long [running deep learning code] takes … because nobody likes to sit there and wait for it to finish” — Karmel Allison at Google I/O 2019

You know the story — you run the code, head out for lunch, and come back hoping your code finished running without breaking. Nodifly ends this process of guessing and checking when the code finishes.

Nodifly is a no-frills, software as a service that can be used to notify users via email or text when the code has been completed, errored out, or met any user-defined conditions. Furthermore, Nodifly provides interactive dashboards to analyze slowdowns, breakdowns, and trends within the coding process.

Current notifications weren’t designed for programmers and are clunky, counterintuitive, and take a long time to embed into the work process. But since Nodifly was created by developers in response to the long run times, it was designed specifically to be as least intrusive, have the least complicated payment plans, and easiest implementation.

2. How do I use it?

From experience, we’ve boiled down the two times notifications are most important: (1) when the entire script finishes from top to bottom and (2) when we want to conditionally notify when a requirement has been reached. Nodifly answers these two occurrences with our standard nodifly.nodifly() which notifies you when the entire script is done, and nodifly.alert() which can be embedded within a process. And we also know how confusing it can be with what code was run and when; so nodifly takes in two parameters: (a) whether you want notifications via text or email and (b) your personal alias for the current process running.

More specifically, Nodifly works on Python platforms whether that’s Jupyter Notebooks or terminal. It’s as easy as importing your standard package and authorizing it with an access key and secret as other API packages.

On Jupyter, it’s as simple as importing the package, authenticating, and calling nodifly.nodifly(kind=’text/email/record’, alias=’your_alias’). The total lines that it takes: 3 lines:

Nodifly for Jupyter

On Terminal, there are two methods depending on if you’re using nodifly alerts or nodifly standard.

Nodifly for Terminal

And that’s it! You can then go onto the Nodifly dashboard tab to view personal dynamically generated interactive dashboards:

Dynamically Generated Interactive Personal Dashboards

3. Where and how do I sign up?

Nodifly was built to work out of the box. After registering for Nodifly (nodifly.com), head to the payments tab and choose one of the basic ($9) or premium ($15) monthly plans. With the basic plan, you have access to 300 texts and 150 emails per month; with the premium plan, you have access to 1000 texts and 500 emails per month. P.S. use ‘medium10’ for a 10% discount off your first months Nodifly subscription at the registration page!

Nodifly Basic and Premium Plans — use code ‘medium10’ for 10% off the first month

After payments have been processed, head over to the settings tab and get the access key and access secret. With these two bits, you’ll have full access to Nodifly text, email, and dashboard services.

Nodifly Settings — With highlighted access key and token for use (keys have already been regenerated :D)

4. What’s in the future for Nodifly?

That being said, Nodifly is always looking to grow into other languages (SQL / R / Java / Spark) and improve functionalities. If you have any questions / comments / concerns / enterprise inquiries, please feel free to contact us at nodiflycontact@gmail.com.

Thank you for reading and please feel free to use “medium10” on sign up for 10% off your first month of Nodifly!

Use ‘medium10’ for 10% off your first month of Nodifly!

Happy Coding,

The Nodifly Team

References:

  1. Machine Learning Zero to Hero (Google I/O’19)
  2. HBR Data Scientist

3. Nodifly.com

4. Nodifly Github Release

5. Nodifly Pypi Official

Stephen Hsu

Written by

Freelance Data Scientist, Machine Learning Enthusiast, I 💖APIs; https://www.linkedin.com/in/stephenjhsu/