After two years of working on Astrobase, through various product iterations, launches, customer interviews, and more, I wanted to share my learnings and my thoughts on what the future holds for me and Astrobase.

Entrepreneurial Harmony

Deploy a private ethereum blockchain on Kubernetes.

Run Ethereum on Kubernetes

Your family and your team are not the same thing.

Full House

What is the last project or business you owned? What did developing that project or business mean for you, for those who helped you build it, for your own learning and self improvement?

Work and Ownership

It takes too much time and energy to be negative.

The Power of Positivity

I think that the goodness of the products and services you build as an entrepreneur can be measured by two factors; how unique and valuable something is.

Unique and Valuable

Hindsight is 2020

The past ten months have felt like a couple of years – so I've decided to stop, reflect, and share some of my learnings.

Trust the Process

The rise in usage of Kubernetes for AI/ML workloads has led engineers to question the suitability and effectiveness of Kubernetes' resource management and scheduling to meet the growing requirements of these workloads. So what?

AI Infrastructure on Kubernetes

If everyone says it isn't working, might as well know how it's supposed to work.

How DNS Works

Many internet services use OAuth and email based protocols for sending login links and tokens. I think it's a pretty slick way to log users in and reduces overhead for them by adding yet another credential to their password managers. Let's walk through how we can implement this – we'll also see a reference to pypale, a package that implements passwordless functionality for us.

Passwordless Logins in Python

Let's run though a text processing exercise in Ruby.

Text Processing in Ruby

Dispatch tables in python are awesome – use them!

Dispatch Tables in Python

Sampling gin, the golang web framework, for building high performance, well curated, production backend apis.

Build REST APIs with Go and Gin

Flutter and Rails from Scratch

Linear regression is fun and easy to do in Go. You don't need to learn python to do basic data work!

Linear Regression with Go

Writing partial derivatives is a great way to understand some of the underlying features of machine learning and neural network libraries.

Partial Derivatives & Recursive Descent

Ever wonder how the heck music appears on your daily drive Spotify playlist? Or how artists and songwriters get paid? Read on to learn about that and how it should be a better system.

Rebuilding the Music Industry

Machine learning engineering is relatively fun and easy to do in Go. Data scientists and ML engineers should checkout Go for building ML applications!

Machine Learning with GoLearn

Spago is an incredibly friendly and effective framework for doing NLP in Go.

Question Answering NLP in Go

Building lightweight ML applications with python, pandas, streamlit, and scikit-learn is a awesome. Let's run through a simple example app to illustrate this.

ML with StreamLit

A great practice in software development is clear error message communication. Pydantic and FastAPI provide an amazing, minimal, intuitive framework for building data based web APIs that make data retrieval a breeze and provide a simple interface for delivering information from client to server.

Clean Rest API Error Messages with Python

Fullstack Kubernetes

Let's build a simple currency converter api with python, FastAPI, and pydantic.

Python Currency Converter

What a year. This is my yearly reflection post for 2019.

Reflections: 2019

Quality of Work

Building a triple flywheel for startup life.

Giving & Growth

What is something that you think is true that most people disagree with you about?

Agreeing to Disagree

After a semester teaching at Carnegie Mellon University's Computer Science department, I wanted to reflect on what it means to teach people better, how to inspire teachers to be better, and how to reorient conversations from machine learning to human learning. I think teaching people should be prioritized over teaching machines.