Contact Info
Tech Stack and Skills
Preferred Stack
#Ryan's 2022 Preferred Tech Stack
#Ryan uses the Anaconda Data Science stack
class ryansPreferredTechStack:
def __init__(self):
self.language = 'Python'
self.ML_lib = 'sci-kit learn'
self.DL_lib = 'Tensorflow'
self.data_tool = 'pandas'
self.cloud_service = 'AWS'
self.web_dev_tool = 'Flask'
self.reporting_tool = 'LaTeX'
self.data_viz = 'seaborn'
self.container_tool = 'Docker'Total Tech Stack
Data Science and Engineering:
➤ Cloud Architecture
- Amazon Web Services (AWS): EC2, CloudFront, Lightsail, Lambda, Glue, Step Functions, CloudWatch, IAM, S3, Route 53, RDS
- Azure: Azure Functions, Function Apps, Azure Storage
- Google Cloud Platform: Cloud Instance, Cloud Scheduler, Cloud Function, Cloud Storage
➤ Data Preparation – nltk, opencv, Labelbox
➤ Data Visualization – Seaborn, Matplotlib, Excel
➤ Modelling Tools – Keras, Tensorflow, scikit-learn, scipy, GPT, YOLO
➤ Statistics – A/B Testing, Hypothesis Testing, T-Test, Bayesian Inference
Software Development
➤ Database Systems – DynamoDB, PostgreSQL, SQLite, MySQL
➤ Programming and Scripting Languages – Python, Bash, R, Docker
➤ Web Development – HTML, CSS, JavaScript, WordPress, Flask, nginx
➤ Development Tools – Visual Studio Code, nano, Jupyter Notebook, Linux, macOS, Vim
Day-to-Day Business Tasks:
➤ Project Management Tools – Toggl, Todoist, Asana
➤ Communication Tools – Zoom, Discord, Slack, Teams
➤ Documentation Tools – Atlassian, Notion, LaTeX, Markdown
I am a Data Scientist with 4+ years experience and an MS in Information Science.
As a kid, I always wanted to work with data. Before kindergarten, I would sit on my father’s lap and read the newspaper with breakfast in the morning. I couldn’t really read words, but I could remember numbers — really well. So much so that the first thing my teachers told my parents was about my ability to recite the morning sports scores when I walked into the room.
My curiosity with data never really subsided since then. In my high school student council, I built the first student-ran school-wide surveys that addressed relevant issues like dress code and assembly participation. During my early years of university, I built an algorithm to take data from different study abroad options to help me decide where would be a good fit. And based on how happy I was the whole time, I would say the algorithm worked!
When I neared the end of my undergraduate studies, I earned an opportunity to apply my data science skills at a cannabis start-up called 4Front Ventures in the summer of 2018. Even though 4Front was already a top operator in the legal industry, I used Python to crunch massive datasets to assist their entry into new markets and perform ad-hoc analyses, further positioning themselves as industry pioneers.
But my biggest work for 4Front did not involve market-level datasets or one-off reports. Instead, I used cluster analysis and innovative data methods to transform transaction data into meaningful consumer segments. What make these segments special is they ditch traditional demographic features for cannabis-specific ones that describe the purchase behavior of a consumer. By swapping out factors like gender and age for flower and edible consumption raters, consumers are receiving more relevant marketing and reducing 4Front’s overall budget. However, perhaps the most important part of this project is it introduced open-source software and analytics to 4Front for pragmatic use. Now, data science thinking and open-source software are some of the first thought-of solutions for many in-house problems
Since 2018 , I have expanded my consulting to include data engineering tasks, web development, and even podcast hosting. In addition to providing data science work in the cannabis industry, I’ve developed cloud infrastructure to improve cultivation and processing data flow and analysis at scale using Google Cloud Platform. Personally, there are few more satisfying things in data science than helping a client turn messy, on-premise, and inaccessible resources into online, scrubbed, and actionable ones that drive real operational decisions.
Outside of cannabis, I am the official webmaster of the Montana 3000 podcast, where I helped up-and-coming author Sean Gallagher present his stories in front of thousands without breaking the bank on hosting or website costs. Sean and I are building a new brand of story-telling from scratch, hoping to disrupt the podcast game with a perfect podcast voice, next-level sound quality, and stories that will make you want to listen twice. Montana 3000 is available on your favorite podcast outlets or at http://montana3000.com
As of now, I currently reside in Arizona, USA and am coping with the pandemic with dozens of houseplants. Check some of them out on a school blog I generated, http://info575papetti.casa
Ryan Papetti 