Topics will include:

• Vector search

• Really big graphs

• Sparse matrices

• Accelerated random forest and XGBoost

• RAPIDS Single Cell and related tools

For the latest updates on location/timing please see the event page at meetup.com:

Event Details at Meetup.com

About the talk: Check out drop in libraries and recipes for everything from rapid data access to traditional machine learning to model tuning! We’ll use the biological sciences as an example, but many of these tools are applicable to any data scientist!

About the speaker: Ben Busby is a Global Alliances Manager at NVIDIA where he helps medical research companies apply advanced AI & ML. He also serves as Adjunct Faculty at Carnegie Mellon University in the Computational Biology Department.

An Unexpected Journey: Life Lessons from Conquering Tesla Solar with Python

Check out the Meetup Event page for all information about time and location.

RSVP and Details on Meetup

About the Talk

In 2024, Christopher installed a Tesla solar array expecting clean energy and lower bills. Instead, he got a masterclass in persistence, reverse engineering, and why you should never trust a system you can’t monitor. Bringing laptops is encouraged for a live interactive surprise at the start of the session!

Tesla’s Powerwall/Inverter ecosystem is powerful but notoriously opaque—out of the box it tells you very little about what’s happening on your roof. When a weird beeping noise caused Christopher to suspect something was wrong, he built his own monitoring solution with PyPowerwall (an open-source library) utilizing Grafana, InfluxDB, and a Raspberry Pi. The shocking result? He discovered hundreds of kWh in lost production annually.

About the Speaker

Christopher Pitstick

In 2024, Christopher installed solar panels on his home and quickly dove into system analysis to optimize performance — a hands-on experience that directly inspired this presentation.

Originally from Dayton, Ohio, Christopher is a Senior Software Engineer at Latitude AI (a Ford subsidiary). His programming journey began at age 12 with QBasic, sparking a lifelong passion that has taken him through roles at Microsoft, Amazon, and Argo AI. He’s worked across numerous languages including C++, Java, C#, TypeScript, Perl, and, of course, Python!

A dedicated neurodiversity advocate, he regularly shares his experiences through public speaking engagements, raising awareness and empowering others in the tech community.

Come join us Thursday, March 26th at CoLab18 in Nova Place for An Unexpected Journey: Life Lessons from Conquering Tesla Solar with Python with Christopher Pitstick (Latitude AI).

After installing a Tesla solar system in 2024, Christopher noticed a strange beeping noise and discovered the system provides almost no visibility into what’s happening on his roof. So he built his own monitoring stack using Python, PyPowerwall, Grafana, InfluxDB, and a Raspberry Pi—and uncovered hundreds of kWh of lost production annually. Doors open 6:00 pm, talk starts at 6:30 pm.

Bring a laptop for a live interactive surprise at the start of the event!

Details and RSVP on Meetup

Sincerely,
~The PyData Pittsburgh Organizers

Due to severe winter weather forecast for the upcoming week, and our desire to keep safety and comfort considerations foremost for our community, we have decided to reschedule this event. This event is now scheduled for Thursday, March 26th. Same time, same location. Please see the event listing for all details.

Thank you, and we look forward to seeing you for the Unexpected Journey, in March!

Sincerely,

~Doug & Madi

Join PyData Pittsburgh tonight at the Swartz Center for our November event — Bodo DataFrames: A Fast and Scalable HPC-Based Drop-In Replacement for Pandas. Ehsan Totoni, CTO and Co-Founder of Bodo.ai, will discuss how Bodo DataFrames brings high-performance computing (HPC) techniques like MPI and JIT compilation to the familiar Pandas API—allowing data scientists to scale Python workloads from millions to billions of rows without rewriting their code.

Details and RSVP

About the talk:
Pandas is a popular library for data scientists but it struggles with large datasets; programs either become too slow or run out of memory. In this talk, we introduce Bodo DataFrames (https://github.com/bodo-ai/Bodo) as a drop-in replacement for the Pandas library that uses high performance computing (HPC) based techniques such as Message Passing Interface (MPI) and JIT compilation for acceleration and scaling. We give an overview of its architecture and explain how it avoids the problems of Pandas (while keeping user code the same), go over concrete examples, and finally discuss current limitations. This talk is for Pandas users who would like to run their code on larger data while avoiding frustrating code rewrites to other APIs. Basic knowledge of Pandas and Python is recommended.

Time:
5:30pm – Doors Open
6:00pm – Talk, Bodo DataFrames: A Fast and Scalable HPC-Based Drop-In Replacement for Pandas

Driving Materials Innovation with Data: The MDS-Rely Center for Industry-Academic Partnerships

Checkout the Meetup Event page for up-to-date information about the time/location.

Meetup Event Page

about the talk

The Materials Data Science for Reliability and Degradation (MDS-Rely) Center is a National Science Foundation (NSF) Industry-University Cooperative Research Center led by the University of Pittsburgh, Case Western Reserve University, and Carnegie Mellon University. MDS-Rely brings together industry, government, and academic partners to conduct pre-competitive research that leverages data science to enhance materials performance, reliability, and service life.

From predictive models for battery degradation to data-driven optimization in additive manufacturing and advanced coatings, MDS-Rely tackles real-world industrial challenges with cutting-edge analytics and experimentation. This talk will provide an overview of the Center’s research focus areas, industry engagement opportunities, and how companies can benefit from and shape this collaborative innovation ecosystem.

MDS-Rely homepage:

https://mds-rely.org/

MDS-Rely on LinkedIn: https://www.linkedin.com/company/mds-rely

about the speaker

Satish Iyengar: I have been at Pitt since I earned my PhD in Statistics at Stanford. My early work was on the approximation of multivariate probabilities and meta-analysis. My interests include applications of diffusion and point processes and multivariate mixture models, especially to problems in psychiatry and neuroscience. I have served on the Food and Drug Administration’s Advisory Committee for Psychopharmacologic Drugs, and for many years I served on the National Security Agency’s Statistics Advisory Group at Stanford. I have long been interested in stochastic models for degradation (many lead to diffusion processes), so I was very happy when Paul Leu contacted me about MDS-Rely in its planning stages.

We are hosting our September event tonight at the Swartz Center, and we would love to see you there!

🎙️ Talk: Decoding Spatial Biology with Python: Multi-Modal Insights into Breast Cancer Progression
🗓️ Tuesday, September 30
🕕 5:30 PM Doors Open | 6:00 PM Talk Begins
📍 Swartz Center for Entrepreneurship, Tepper School of Business, CMU

Free Parking is available after 5pm in the Morewood Gardens Lot, directly behind the Tepper School building. The entrance is off of Morewood Avenue.

Join us as Alex C. Chang, CMU-Pitt (Graduate Student PhD, Computational Biology) and Brent Schlegel, University of Pittsburgh School of Medicine (Graduate Student PhD, Integrative Systems Biology), present their novel tool, CITEgeist, which harnesses Python’s capabilities for multi-modal spatial transcriptomics.

Details and RSVP on Meetup

We look forward to seeing everyone at tonight’s Python for All: Democratizing Coding Mastery with AI Chatbot Support event. Kristen Scotti, CMU STEM Librarian, will share her interactive workshop, where we will explore how AI chatbots can be invaluable resources for understanding, writing, and refining code, as well as for facilitating transitions between different coding languages.

Details and RSVP on Meetup

July PyData Pittsburgh Meetup is happening tonight – and we’d love to see you there!

🗓️ Wednesday, July 30
🕕 6:00 PM Doors Open | 6:30 PM Talk Begins
📍 Giant Eagle Auditorium, Baker Hall, Carnegie Mellon University

(Follow the walkway to the right of the building on Frew Street and head down the steps to the basement entrance – [photo and directions below])

Tonight’s Talk: Does Generative AI Know Statistics?

There’s a lot of buzz about Gen AI, but what are some scientifically grounded ways to interact with it?
Join us as Dr. Louis Luangkesorn explores how Generative AI can be a partner in statistical analysis—warts, randomness, hallucinations, and all.

Preview the talk: Dr. Luangkesorn has shared a Jupyter Notebook here

About the Speaker

Dr. Louis Luangkesorn is a Lead Data Scientist at Highmark Health working at the intersection of statistical modeling, operations research, and Generative AI in healthcare and HR. He contributes to SciPy and SimPy and co-authored a chapter in Data Science in Health Informatics: An Interprofessional Approach. He is also VP of INFORMS-Pittsburgh.

About PyData Pittsburgh

We’re a community of data scientists, engineers, and researchers passionate about Python, data, and science. Our monthly events are a space to connect, learn, and share ideas.
Questions or want to get involved? Reach us at organizers@pypgh.org

Getting to the Venue

• Free parking along Frew Street after 6PM and at Morewood Gardens Parking Lot after 5pm.

• Head to Baker Hall, CMU

• From Frew Street, take the walkway on the right (east) side of the building

• You’ll find steps leading to the basement entrance

• The Giant Eagle Auditorium is just inside
  

  

  

  

We're excited to invite you to our next meetup, where we'll explore the intersection of Generative AI and statistical reasoning. There's a lot of buzz about "Gen AI", but what are some scientifically-based ways we can interact with it? Join us on Wednesday, July 30 for our July Meetup: Does Generative AI Know Statistics? as Dr. Louis Luangkesorn presents how Generative AI can be useful as a partner in analysis.

More information about the event can be found on our Meetup event page!

About the speaker:

Dr. Louis Luangkesorn is a Lead Data Scientist at Highmark Health where he works on projects applying statistical, predictive, operations research, and Generative AI models in use cases involving human resources and healthcare. He has contributed code to SciPy and SimPy as well as the book chapter on Data Science in Health Informatics: An Interprofessional Approach. He is the current VP of INFORMS-Pittsburgh, the Pittsburgh chapter of INFORMS.

About PyData Pittsburgh:

PyData Pittsburgh is a local community of researchers, engineers, and data enthusiasts who share an interest in Python, data, and science. We host monthly meetup groups where community members can give a talk about the exciting work they've been doing. We aim for these events to foster collaboration, learning, and connection in the Pittsburgh data ecosystem.

Want to get involved or have questions? Reach out to us at organizers@pypgh.org!

📣Subscribe to our Meetup Page

📬Join our Substack

👉Follow us on LinkedIn

Happy Juneteenth!

We hope you’ll share some of the holiday with us this evening at PyData Pittsburgh’s June event: Panning for Gold in the Streams of Conversation.

Join us at COhatch Waterfront as Jay Palat, Head of AI/ML at Mpathic, walks us through how machine learning and Python tools are used to analyze these conversations and turn them into actionable insight.

Come learn more about this exciting topic!

Special thanks to COhatch for hosting us at their beautiful Waterfront location.

Details and RSVP in Meetup

Today is the day! If you’re at PyCon US 2025, join us this afternoon for the Hometown Heroes program, a track featuring eight talks from all Pittsburgh-based speakers.

🎤 Hometown Heroes Program
Saturday, May 17, 1:45–5:30 PM
Room 317
David L. Lawrence Convention Center

View the Hometown Heroes Schedule

Later this evening, join us for the PyData at PyCon Happy Hour at Aslin Beer Company in the Strip District. We’ll be welcoming both members of the local Python community and PyCon attendees visiting Pittsburgh from around the world. The Happy Hour is free and open to the public — you don’t need to be registered for PyCon to attend!

🎉 PyData at PyCon Happy Hour
Saturday, May 17, 6:30 PM
Aslin Beer Company

RSVP for PyData at PyCon Happy Hour

Hope to see you there!

Hey all ~

We’re putting together a PyData Pittsburgh community booth in the Expo Hall this year, and we’d love to invite you to help us staff it for a shift. It’s a great way to meet other Python folks, share what PyData Pittsburgh is all about, and get more people excited about the local community.

Go to the sign-up form

You’d be helping us out a ton — and connecting with folks who might want to join or support PyData Pittsburgh. We’re hoping to grow the community with help from amazing local folks like you.

Thanks so much for considering!

And of course, if you just want to stop by our booth and say hello, we would love that, too!

Warm regards,

~Doug & Madison

Hi all!

Just a quick reminder that the deadline for submitting a proposal for the Hometown Heroes track at PyCon 2025 is coming up soon! You have until Sunday, April 6th, at 11:59pm to send us your proposal.

Go to the Call For Proposals Form

We are seeking proposals for 30-minute presentations and 5-minute lightning talks that highlight interesting and innovative uses of Python. This is a great chance to connect with our local tech community, and to share your insights at the biggest Python conference in the world!

The Hometown Heroes track will take place on Saturday, May 17, from 1:45 PM to 5:45 PM, on-site at PyCon US 2025 at the David L. Lawrence Convention Center.

We look forward to receiving your proposal, and to seeing you at PyCon!

Cheers!
Doug, Madison, Patrick, & Nathan
Organizers, PyData Pittsburgh

We look forward to seeing everyone at tonight’s Machine Learning in Astronomy event at the Allegheny Observatory! Ashod Khederlarian, a 4th-year Ph.D. student at the University of Pittsburgh, will share how state-of-the-art machine learning techniques are used to analyze vast astronomical datasets.

After the presentation, staff members will offer a free private tour of the observatory exclusively for the PyData Pittsburgh group. See you then!

Details and RSVP on Meetup

There are only two days left to submit proposals for talks, tutorials, posters, and Charlas at PyCon US 2025 in Pittsburgh! The Call for Proposals closes at the end of the day on Thursday, December 19. If you’ve been thinking about submitting something, now is the time!

PyCon US 2025 Call for Proposals

If you do submit a proposal to PyCon US 2025 — consider submitting it to the PyData Pittsburgh Call for Proposals, too! We’re always looking for new speakers, and we could provide an opportunity for you to road-test and refine your material in front of a friendly audience before the main conference.

We’re looking forward to seeing everyone at tonight’s Data science for mental health crises event featuring Pim Welle, Chief Data Scientist at the Allegheny County Department of Human Services! We’ll also have several announcements to share about major upcoming events of interest to the Python, machine learning, and scientific computing community here in Pittsburgh.

RSVP on Meetup

On a bittersweet note, this will be our final event hosted at the Code & Supply Workspace and Community Center, which will be shutting down in December. (Code & Supply’s online community and other activities will continue as usual.) We’ve always moved our PyData Pittsburgh events around to different locations, but the Code & Supply Workspace has been a staple in our rotation. Let’s show our appreciation to Colin Dean and the rest of the team at Code & Supply by packing the house one more time.

RSVP on Meetup

We’re at 79 RSVPs and counting, so you’ll be in good company.

We’ll be gathering in person at the Swartz Center for Entrepreneurship at CMU. Food and drinks will be provided. See you there!

We’ll be gathering in person at the Swartz Center for Entrepreneurship at Carnegie Mellon University. Food and beverages will be provided.

For more information about Rorry’s talk and to RSVP, please visit the event page on Meetup.com:

RSVP on Meetup

We have 51 RSVPs so far. I’d love to hit 75!

Want to come up to speed on neural networks?

Rorry’s talk will be about the inner workings of neural networks and how to improve them. Want a concise, intuitive introduction to the current generation of neural networks and how they work? Or has it been a while, and it would be helpful to refresh your memory?

The Neural Networks series from 3Blue1Brown is one of my favorite resources on the topic:

The first five videos — focused on the structure of neural networks and the backpropagation algorithm — will be the most relevant to next week’s event. (The last two videos provide an excellent introduction to the transformer architecture and the attention mechanism, but that’s a topic for another day.)

Help us spread the word!

Know any friends or colleagues who might be interested in Radically improving neural networks with insights from modern neuroscience or our other events? Tell them about PyData Pittsburgh! Word of mouth is by far the most effective way new folks discover our community.

You can also feel free to re-share these posts about our June event on your favorite online platforms — that helps too:

• LinkedIn: https://www.linkedin.com/feed/update/urn:li:activity:7201269410274123776/

• Mastodon: https://fosstodon.org/@PyDataPGH/112519876670704872

• Twitter: https://x.com/PyDataPGH/status/1795505498734649445

• Web URL: https://news.pypgh.org/p/radically-improving-neural-networks