About Me

Ever since I was young I have loved to build things. But my love for building machines that could help me solve problems started with the simple game Mancala. After sifting through Game Theory papers and my first Artificial Intelligence literature, I built a game-tree solver for that simple 2-player African board game. Further cementing my interest in artificially intelligent systems were projects predicting the 2019 NFL Hall of Fame awardees, and developing a computer vision app to solve the simple question of "Is this compostable or recyclable or neither?!"

Since these experiences, I've decided to build out more impactful artificially intelligent systems at scale. Most recently, I'm very interested in building on top of ₿itcoin to get this open financial network in the hands of those who need it most internationally. You may explore my formal work experience here on this site.

What I'm Doing

Algorithms & Data Structures

I've been learning efficient systems since I was 16 and purposely brush up on these daily.

Machine Learning Operations

Building out machine learning pipelines (Kafka, Spark, Scala) and infrastructure (ETL with Python, SQL, AWS Lambda).

Backend Development

Professional, fully OpenAPI compliant API development and application engineering in Java, Scala, and Python. Well versed with SQL and NoSQL document based databases.

Artificial Intelligence Modeling

I've built out productionized machine learning models for Capital One for real-time fraud detection, insider threat detection, command and control detection, and my latest project is NLP and violence mitigation.

Testimonials

Vannia Gonzalez - Senior Data Scientist

I think Michael is a very good engineer and also has good analytical skiils, which are two sought-out skills for an ML engineer. More importantly, he was excited to learn and very eager to understand every aspect of the project, from the engineering backend to the model behind the workflow so he could meaningfully contribute to the project. He did a good job getting up to speed and I have no doubt he will do a great job contributing to the product in the following versions and enhancements. I think it is also important to highlight that he reached out based on his own interest and eagerness to learn more about the model validation stage, anomaly detection, and time series.

Gurpreet - Software Engineer

I think you've got a great sense of ownership of your work. I like that you firmly believe in following programming/teamwide best practices and you walk the walk. You always offer help to the teammates. One Instance I can easily remember is when I had recently joined the team you completely demystified the SQL for me. Your opinions are always backed by sound reasoning.

Lee Adcock - Director

He's shown himself to be an effective communicator and relentless problem solver tracking down challenges and iterating towards continuously improving solutions. Specific competencies that stand out as key strengths for Michael include his ownership and his reputation as a strong communicator among a variety of audiences.

Michael Loy - Scrum Master

I'm really encouraged by Michael's depth of curiosity. Lots of tasks assigned to new engineers can be seen as general "business as usual" activities; Michael approaches even the simplest task with thoughts on potential improvements. He never looks at simply finishing requirements - he nearly always adds his own expertise and research into his work. The curiosity will take him far, and I always hope he continues following this approach through the next stages of his career.

Patrick Sofo - Product Owner

After only a few months on the team, Michael has helped with multiple high priority initiatives. Michael is very eager to jump right into new problem spaces and learn. He is also open to feedback from the team on his own work and actively elicits this feedback by effectively communicating progress and questions as he goes.

Devin Rose - Senior Engineer

Michael implemented a vital diagnostic tool for the back-end of our web application that enables us to quickly and intuitively search our enormous robot logs to troubleshoot problems in our system or with robots. He used cutting edge streaming technologies to efficiently pipe the log messages into Elastic Search in AWS, intelligently indexed the data for our search needs, and exposed the functionality via Kibana.

Where I've Been

Resume

Education

Rice University

2015 — 2019

Bachelor of Arts - Computer Science and Concentration in Business

The Gatton Academy @ Western Kentucky University

2013 — 2015

Enrolled in Western Kentucky University Honors program at age 16. Here is where I initially found programming.

Experience

Machine Learning Engineer - Capital One

2020 — Present

• Overhauled serial ETL (Extract, Transform, Load) processes into a concurrent design leading to miniscule computational footprint – optimized from 12 hours to 16 minutes AWS EC2 compute time (45x speed-up) $200,000/year saved

• Built Snowflake, SQS, Lambda, Kafka machine learning pipeline for real-time insider threat detection of AWS logs to combat. Utilized model selection to fit data with seasonal, trend, or serial models respective to each subset’s features

• Implemented Scala, Spark, AWS Redshift machine learning pipeline to predict real-time fraud

• Identified a flaw in data cleaning, increased the healthy data from 50% to 90% useable data (millions of records)

• Discovered OWASP Top 10 vulnerability for AWS serverless implementation – NPI over DNS Lambda exfiltration

• Explained beacon detection model and findings and security vulnerability to Senior Director

• Designed, implemented, & documented AWS Lambda driven entire REST API compliant with OpenAPI standards

Chief Technical Officer - RoyalBooks

2019 — 2020

• Translated engineering concepts & results into business outcomes for CEO

• Scoped engineering plans from business goals

• Designed & PoC’d system architecture: Firebase Cloudstore connected to Flutter/Dart for hybrid mobile development

• Enabled junior engineers by creating tutorials, NoSQL schemas, UI diagrams, and documentation for coding best practices

• Mentored 4 new grad engineers & delegated development by skillset.

• Led team in scrum meetings, sprint planning, and retrospectives. Interviewed technical candidates

• Reviewed code & implemented unit testing framework

Founder - Wireless Internet Service Provider

2020 - Present

• Built 700-2500 MHz cell receiver site with parabolic and LDPA antennas, optimized cellular modems (TTL, UID spoofing) with PoE copper ethernet backhaul, leading to a mesh node PTMP subnet distributing internet to rural community

• Boosted network performance ~20x (~1000ms, 8mbps, and 0.5mbps) to (~50ms ping, 130mbps, 50mbps)

Skills

Amazon Web Services & Serverless

ETL Pipelines

Concurrent and Parallelized Programs

Clear Communication & Ownership

Languages

Python and SkLearn

SQL and Snowflake

Scala

Java

Portfolio

Select category

Category
All
Concept
Design
Life

Blog

Sep 16, 2019

Sample Post 1

Veritatis et quasi architecto beatae vitae dicta sunt, explicabo.

Sep 15, 2019

Sample Post 2

Sed ut perspiciatis, nam libero tempore, cum soluta nobis est eligendi.

Sep 14, 2019

Sample Post 3

Excepteur sint occaecat cupidatat no proident, quis nostrum exercitationem ullam corporis suscipit.

Sep 13, 2019

Sample Post 4

Enim ad minim veniam, consectetur adipiscing elit, quis nostrud exercitation ullamco laboris nisi.

Sep 12, 2019

Sample Post 5

Maxime placeat, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua.