Talks

These are the talks accepted to PyTexas, listed in no particular order. A full schedule will be released close to the start of the conference.

Verbs, Not Nouns: Writing Documentation Users Want to Read

Lots of documentation is dull and tough to read. That's because most documentation focuses on describing product features, and those are nouns. Your user's primary concern isn't what your product is, or how it's made. Your user has a problem they want to solve, and they want to know how to use your product to do that. They're looking for a verb. In this talk, you'll find the verbs that go with your nouns, and how that translates into readable documentation.

Speaker: Brian MacDonald

Brian MacDonald has been an editor of technical publications for over 30 years. Currently, he's a Technical Content Engineer for Temporal. He's worked on community documentation for DigitalOcean. In the past, he provided freelance services to technical publishers, with clients including O'Reilly, Pragmatic, Wiley, Apress, Wrox, Osborne, and Manning. He has co-authored two editions of Learning C# and Learning ASP.NET for O'Reilly. You can follow him on Mastodon at @[email protected]. He lives outside of Philadelphia, and enjoys riding roller coasters.

The Pythonic Ideal in the Age of Generative AI

The advent of generative Artificial Intelligence (AI) programs, such as ChatGPT, and code pair programming tools, such as GitHub Copilot, has ushered in concerns about whether or not it is still valuable to learn to write code, or to continue to improve our ability to write idiomatic, beautiful code. The reality is that generative AI programs can write code that is poor quality or incorrect. In addition, generative AI is still in its infancy and cannot yet replace human innovation or innate expertise for complex tasks, such as writing Pythonic code. Do we need to be worried about AI taking over coding jobs? This talk will cover why learning to write Pythonic code is still a valuable skill to pursue, and will introduce techniques to add generative AI and pair programming AI tools into your teaching (and learning!).

Speaker: Heather Crawford

Heather Crawford is an Engineering Technical Trainer at Bloomberg and a former university professor. In her roles as a professional educator, she has taught learners at many levels from university students at all levels to both entry-level and experienced software engineers. Heather is a Python lover, who has used it since 2007.

Typed and Decorated Python GUIs: Help the Language Help You

Python's increasingly powerful type annotations (and the type checkers and IDEs that consume them) are a great way to manage the complexities of Python GUI programming. This talk will walk through a strictly typed PyQt/PySide GUI application skeleton while touching on custom logging handlers, enum-based error passing, type-aware match statements, and how generic types empower function decorators, decorator factories, and class decorators.

Speaker: George Collins

I'm a practicing attorney (senior counsel at Phillips and Cohen LLP) who represents whistleblowers in False Claims Act litigation as well as the SEC, CFTC and related whistleblower programs. I specialize in evidentiary data science, using Python (plus a handful of other languages) to comprehensively understand data sets obtained through discovery from the companies that our clients are blowing the whistle on. Without Python (and the libraries, tools and community that have grown up around it) my work would be considerably slower, far lonelier, and much less satisfying!

Introduction to OpenTelemetry with Python

OpenTelemetry (OTel) is a comprehensive suite of APIs, libraries, and tools designed to collect, generate, and export telemetry data, including metrics, logs, and traces, from software applications. This talk will cover critical concepts related to OTel and show you how to integrate it into your Python application.

Speaker: Jessica Garson

Jessica Garson is a Python programmer, educator, and artist. She currently works at Elastic as a Senior Developer Advocate. Previously, Jessica was at Twitter for four years, working in Developer Relations. She has spoken at conferences all over the globe, ranging from PyCon to Write the Docs. In her spare time, she uses code and modular synthesizers to make music and audio-reactive video art.

Democratizing Data: Python-Powered Dashboards and Open Data for Transparent Governance in Austin

Governments generate vast amounts of data, but making it accessible, actionable, and engaging for the public remains a challenge. This session explores how the City of Austin leverages Python to transform open data into interactive dashboards that foster transparency and empower citizens. Through real-world examples from Austin's public sector projects, attendees will learn how data visualization and analytics can strengthen community engagement, enhance decision-making, and create a more informed society.

Speaker: Tanvi Sharma

Tanvi Sharma is a Gen Z data enthusiast with over 5 years of experience in Data Science, Machine Learning, and Data Visualization. As a data scientist with the City of Austin, she champions open data and builds dashboards that drive transparent governance and community impact. With a LinkedIn following of 12,000+, Tanvi loves sharing insights and inspiring others in the data world. Equal parts techie and storyteller, she's passionate about using data to empower smarter cities and stronger communities. Off the clock, she's an adventurer, culture enthusiast, and a dancer at heart.

Building a test framework from scratch (or not)!

pytest is a fantastic test framework, but why use it when you can just build your own test framework from scratch? Sounds wild, right? Well, it is! In this talk, we will learn what testing in Python is all about by iteratively building features of a framework from the ground up. Through this, we will learn not just what goes into a test framework but why those features are there - the runner, the structure, the assertions, and more. We will develop a deep understanding of how test frameworks work and also how to write good tests with them. We will also gain immense appreciation for mature test frameworks like pytest that save our time by providing off-the-shelf testing features that deliver serious value.

Speaker: Pandy Knight

Andrew Knight, also known as 'Pandy', is the Automation Panda. He's a software quality champion who loves to help people build better quality software. Currently, he works as the Senior Director of Product Management at Cycle Labs, focusing on building an excellent test automation platform for enterprise systems. Previously, Pandy spent a decade as a SDET building solutions to testing problems at various tech companies. He also previously led Developer Relations and Test Automation University at Applitools. As an avid supporter of open source software, Pandy is a Playwright Ambassador as well as the lead developer for Boa Constrictor, the .NET Screenplay Pattern. On the side, he is writing a book on software testing with Manning Publications. Apart from software, Pandy spends time with his family, his French Bulldog, and his vintage Volkswagens. Check out his tech blog at AutomationPanda.com, and follow him on Twitter at @AutomationPanda.

API Design for Those That Don't RTFM (aka everyone)

Speaker: Lynn Root

Your users shouldn't need to go on a scavenger hunt through documentation just to get started using your API. This talk explores how to bridge the gap between usability theory and practical API design, focusing on creating interfaces that are so intuitive that the docs become optional.

Lynn Root is a Staff Engineer at Spotify and an adjunct professor at Columbia University's Graduate School of Engineering. She is a seasoned speaker on building and maintaining distributed systems, and is the tech lead for governance compliance for Spotify's ML/AI platform. Lynn is a global leader of diversity in the Python community, the Chair of the PyLadies Global Council, and a PSF fellow and the former Vice Chair of the PSF's Board of Directors. When her hands are not on a keyboard, they are usually holding a pair of knitting needles or a bass guitar.

Building a distributed asyncio event loop

To support distributed, durable (i.e. resumable) asynchronous code in Temporal, a custom asyncio event loop was required. We will cover how asyncio event loops work under the hood, considerations for building a custom, deterministic one, and gotchas encountered along the way.

Speaker: Chad Retz

Chad Retz is a Senior Staff Software Engineer at Temporal and the developer of Temporal Python used for developing distributed workflows as Python code.

LLMs as Bug Hunters: Leveraging LLMs to Discover Python Vulnerabilities

Large Language Models (LLMs) are changing how we think about software security, making it possible to discover vulnerabilities that evade traditional tools. Melding the powers of AI understanding code semantics with Python's flexibility, we can develop tooling to track complex input/output paths, reveal multi-step exploits, and make code safer. This session will examine how Python is being used to create state-of-the-art vulnerability detection tools that harness LLMs to find new, previously undiscovered security bugs.

Speaker: Jessy Ayala

Jessy Ayala is a third-year PhD student at UC Irvine studying problems in open-source security and how they fit in the software supply chain ecosystem. Previously, Jessy worked as a cybersecurity engineer in the health-tech industry and has extensive experience in security testing, vulnerability triaging, large-scale data analysis for security, and mixed-methods security research.

Place-making and productivity: Build maintainable broad-scale tools with a small team

As we bring on new team members, we want to ensure they can ramp up quickly and have a good experience. They should be able to understand "the company way" of accomplishing things and quickly find the tools they need to do their job. Veteran team members should also be able to codify and commoditize the patterns they've developed over time so they can focus on higher-order problems. In large organizations, such tooling might be the purview of a dedicated platform engineering team and have a budget to match. But what if you're a small team, or a team of one, and you want to build something that scales to the whole organization without becoming a full-time job? At ITHAKA, over 100 engineering staff manage hundreds of applications in a handful of languages and frameworks across many teams and products. Despite the variety, most of those engineers still need to perform a core set of tasks when developing, from debugging to traffic routing to feature toggles. Teams are always working on high-priority projects, and it can be difficult to find the time to take stock of things that can be better streamlined or automated. Further, although we have a platform engineering team, as a non-profit we don't have the budget for someone to work on meta-tooling as a full-time job. Over time, several aspects of our development process have become common sources of friction, with a lot of time spent answering questions or debugging local machine setups. In this talk I'll tell the story of my journey to build a tool with a net positive return on investment, from the initial idea to the rollout and adoption. I'll cover the principles that guided my decisions, the tools we use, and the outcomes we've seen so far.

Speaker: Dane Hillard

Artsy, fartsy, musical computer nerd who likes modular software, biotechnology, education, and metacognition.

Signal Processing in Electrochemistry with Python: Applications to the US Opioids Crisis

With an average of 195 daily deaths due to synthetic opioids overdose in 2021, the US have been facing an unprecedented opioids crisis. Fentanyl and its analogues have been a major source of concern, due to their high levels of addiction, fast-acting mechanisms, and detection challenges. Fast, effective, and accurate identification and quantification of fentanyl, its analogues, and metabolites are essential to help prevent overdose-related incidents and to enable agile medical response. Although electrochemical sensors represent a promising technology for selectively detecting opioids at low concentrations, analyzing and processing the data remains one of the major challenges. To tackle this challenge, certain Python libraries, such as scipy.signal, could be useful in processing signals with high levels of noise and interference from other substances. This talk will focus on specific examples of how such libraries could help enable filtering, Fourier transformation, and wavelet analysis of electrochemical data. As examples, code snippets and outputs will be shown to demonstrate how Python can help improve the quality and usability of electrochemical data. Additionally, this talk will show how data processing can further benefit from hyper parameter optimization techniques, enabled by libraries like hyperopt and ray-tune. By doing so, my hope is to demonstrate how Python could be extremely useful in helping address problems at the intersection of statistics, public health, and public policy.

Speaker: Rodrigo Silva Ferreira

Rodrigo Silva Ferreira was born and raised in Salvador, Brazil. He obtained a BSc. in Chemistry with minors in Applied Math and Arabic from NYU Abu Dhabi, and a MSc. in Analytical Chemistry from the University of Pittsburgh. Rodrigo currently works as a software quality engineer at a statistical software company in central Pennsylvania. When he is not busy testing statistical products, Rodrigo enjoys learning about how statistics and data can be used to develop a deeper understanding of topics at the intersection of STEM, society, and public policy. He is passionate about using statistics and data to understand and improve the world.

Demystifying the World of Serializers (and Deserializers) for SQLAlchemy ORM Models in Python

SQLAlchemy simplifies SQL model creation and manipulation using Python. However, efficient data transmission requires serialization of Python objects for network interactions and database storage. Serialization transforms data for transmission and storage, which are crucial in database procedures. In this talk, we will explore configurable (de)serialization tools to enhance SQLAlchemy ORM model's capabilities, their advanced use cases, and implementation. We will also compare this to the features of commonly-used serializer alternatives, such as SerializerMixin, Marshmallow-SQLAlchemy, ColanderAlchemy, Serialchemy, SQLAthanor, and more.

Speaker: Ritik Mathur

Ritik Mathur is a software engineer in Bloomberg's London Office, where he has been developing with Python for 2.5 years. He works in the field of climate risk analysis, supporting the BloombergNEF engineering team in their goal of building Bloomberg Terminal functions and tools that help clients discover and interact with BloombergNEF's data, news, and research into a sustainable future. He is passionate about utilizing software and data engineering to drive business decisions.

One Year In: Insights and Inspiration from my PSF Journey

Gain insights from my first year at the PSF, exploring how the CPython project, its surrounding efforts, and the PSF operate. Discover how you can contribute - code to documentation - to help shape Python's future.

Speaker: Jacob Coffee

Jacob Coffee is an Infrastructure Engineer at the Python Software Foundation and a newly invited CPython triager. He supports key Python services such as PyPI.org and Python.org while also contributing to the maintenance of the Litestar ecosystem which boasts libraries such as Litestar, Advanced Alchemy, Polyfactory, and more. He is passionate about open-source development, the mission of the PSF, and enhancing the tools that empower developers worldwide.

Python meets the crab - Friendship at first sight?

Python's simplicity often comes at the cost of performance, especially in computationally-intensive tasks. In contrast, Rust enables one to write low-level code that takes full advantage of the underlying hardware. Furthermore, Rust (just like Python) has a vibrant ecosystem with many great libraries. PyO3 (together with the maturin build back-end) allows you to easily integrate the two languages together and make use of the best of both.

Speaker: Bernát Gábor

Over the years, I have delivered 2 presentations + 1 workshop at PyCon US; 1 presentation at PyTexas, 4 presentations at EuroPython, 1 at PyCon UK, and two talk sat PyLondinium. My website (https://bernat.tech/presentations) contains links to the videos and slides for these presentations. I am also the author of the virtualenv/tox packages and a PyPA member.

Python Untethered: Building Robust Embedded Systems

When code finally leaves your IDE, the destination is often not the cloud or a PC, but an embedded device in a home, office, or even the wild. However, developing applications for the embedded space comes with its own unique set of challenges and pitfalls. Whether designing for the home or enterprise, embedded devices require additional considerations for usability, reliability, and maintainability. In this talk, we'll explore real-world challenges of embedded Python development and provide practical solutions drawn from both DIY projects and enterprise deployments. You'll learn strategies for:

• Ensuring system reliability through power failure tolerance and fault handling
• Building secure remote access and upgrade capabilities
• Implementing effective monitoring and observability
• Managing hardware constraints and system resources
• Designing flexible configuration systems that work in the field

Whether you're a hobbyist looking to make your home automation more robust or a professional developer designing industrial systems, you'll walk away with concrete patterns and best practices to build more reliable embedded Python applications.

Speaker: Oliver Rew

Oliver Rew is a software engineer at Bloomberg in New York City, where he works on data center technologies and infrastructure automation. Outside of work, he enjoys electronics, hardware design, and anything related to space (ask him about his CubeSat!).

Deploy Your Next Python App with WebAssembly (Wasm): Smaller, Safer, Faster

This talk will highlight the unique advantages of deploying Python applications using WebAssembly (Wasm) on the server. Moving away from conventional deployment strategies, Wasm introduces a groundbreaking approach, promising smaller, faster, and more secure server applications. We'll explore the integration of Python with server-side Wasm environments, focusing on the performance, security, and architectural innovations it offers.

Speaker: Dan Phillips

Daniel Phillips is a Senior Engineer at Loophole Labs, where one of his main areas of focus is server-side WebAssembly. He is also the founder and organizer of the WebAssembly Chicago group.

Reducing the "Oops Factor": Pipelines for Securing your Python Development Lifecycle on a budget

The cyber threat landscape is vast, deep and ever-changing. Short of retraining as cybersecurity professionals, How can we, as Python developers do our part to help keep ourselves, our customers and our data safe?

In this talk, we'll look at the current threat landscape, the ways developers commonly fall short, and just how cheap, quick and easy it can be to get some quick wins and drastically reduce the "oops factor" of our Python development lifecycle.

Speaker: Simon Merrick

Simon is a Senior Cloud Engineer at Trade Me, with a background in software development with Python. Previously Simon was a contributor to OpenStack, an open source cloud platform written in Python which sparked his joy for Cloud and DevOps. Coming from a software development background Simon approached his cloud Journey like learning Python - through writing code to build and automate and secure cloud infrastructure.