PyConline AU 2021
| title | author | length |
|---|---|---|
| Welcome to PyConline AU 2021 | 02m | |
Opening remarks list concrete logistics and highlights: 50 talks across specialist tracks and lightning sessions, social events including a conference karaoke… | ||
| High performance machine learning with JAX | Mat Kelcey | 31m |
JAX is a Python front end to XLA that gives automatic differentiation and a drop‑in jax.numpy replacement so code often runs on GPUs or TPUs after swapping… | ||
| Fair game: the ethics of telco fraud | Laura Summers | 30m |
An ethics ops consultant embeds with a small telco fraud team using simple deterministic rules plus a third‑party ML payment gateway, chasing international… | ||
| AppSec on a budget | Mario Areias | 29m |
Session gives six concrete tactics for doing AppSec on a shoestring: find cross‑team allies, leverage security incidents to win headcount (an incident wins an… | ||
| Cataloging the world's data for great good | Lars Yencken | 36m |
Our World in Data claims to catalog and republish global datasets with provenance and visualization metadata, keeping an internal MySQL catalog and publishing… | ||
| The Butterfly Effect: How a single bit changed my career | Jon "maddog" Hall | 26m |
A single missing execute bit on a startup script breaks deployment and mail flow on 150+ appliances offline during Black Friday, forcing SSH or phone-guided… | ||
| Python Type hints and how they can supercharge CS education. | Shrey Somaiya | 29m |
Python type hints claim to reduce runtime errors by catching mismatched inputs before execution — the email Message example shows an annotation turning a… | ||
| Planning for Failure using Chaos Engineering | Amit Saha | 31m |
Claim: chaos engineering techniques can continuously assess production readiness by running controlled experiments that prove or disprove concrete hypotheses… | ||
| When a little web app needs to grow up... the journey from a hobby project to production | Matt Coles | 33m |
Starts as a Flask weekend prototype that emails orders to a vendor, grows into a Python app handling hundreds to a few thousand orders, and ships t-shirts… | ||
| Vampires in the Browser: banishing uninvited Javascript from your web app | Lilly Ryan | 22m |
Examples show third‑party JS actually stealing data and computing power: CoinHive cryptojacking mines tiny sums, while Magecart skimming compromises ~400,000… | ||
| Time Series Forecasting Techniques in Python | Gajendra Deshpande | 29m |
A practical overview that claims to teach basic model selection and shows runnable examples: imports from statsmodels, a random toy dataset, fitting… | ||
| horse. | Christopher Neugebauer | 01m |
A comedic thesis claims everything is a horse, born from a 5am insomnia riff and delivered as rapid reclassifications: camel is a lumpy horse, tiger is a… | ||
| A tale of Python C extensions and cross-platform wheels | Vinayak Mehta | 24m |
Creates a C++ wrapper with pybind11, implements a PDF→PNG converter using Poppler, and packages extension modules with setup.py so pip builds a shared-object… | ||
| Knock, knock! Who's there? | Markus Holtermann | 30m |
Covers concrete authentication choices and tradeoffs: password hashing with argon2 and scrypt (and why salts and cost/memory matter), RFC basic/digest and… | ||
| Education Track Closing Address | Amanda J Hogan, Nicky Ringland | 04m |
Closing notes list concrete details: check pycon.org.au/social for tonight’s virtual events, expect sponsor shoutouts and an appeal to praise backers on social… | ||
| Python VS Space Junk | Mars Buttfield-Addison | 33m |
Summarizes how existing astronomy sensors get repurposed to improve orbit predictions and collision warning accuracy, uses astronomy sensors for STM, shows… | ||
| Making Swiss Cheese of Ticket Sales | Yaakov | 30m |
Custom ticketing system built on TypeScript React Node Express MySQL on Kubernetes collapses two minutes into a launch, selling one ticket in the first 100… | ||
| Improved Decision Making with Pareto Fronts | Eyal Kazin | 34m |
Claims that Pareto fronts expose the set of non‑dominated tradeoffs so no precommitted weighting is needed, and that holding subjective choices until after… | ||
| At the Turn of the Tide – Teaching Python in Real-Time at Scale | Stephen Tierney, Alison Wong | 30m |
A university course migrates to a browser‑based learning platform called ed and builds interactive lessons with code challenges, automatic marking, and shared… | ||
| Classifying Audio into Types using Python | Jyotika Singh | 31m |
Uses pi audio processing in Python to extract specific features (MFCC, GFCC, spectral, chroma), train scikit learn SVMs and ship three pretrained classifiers… | ||
| Graph Data Science | Paco Nathan | 30m |
Claims an integration stack rather than a proprietary platform: kglab abstraction layer ties RDF/JSON‑LD and Parquet storage to pandas, NumPy, scikit‑learn… | ||
| Hey, Python-Web-Community! What's going on? | Vibhu Agarwal | 31m |
Explains how Python web pieces actually connect: CGI versus pre-fork workers (fork cost, worker pool), WSGI’s single-call app and server separation, and ASGI’s… | ||
| CFU Playground | Alan Green | 31m |
A practical pipeline for building FPGA ML accelerators with an open source toolchain that actually runs on hardware and speeds up real models: it uses nMigen… | ||
| So You’re a Software Developer, Now What? Exploring Career Growth | Brianna Laugher | 30m |
Argues that career growth is best treated like product‑market fit with a concrete five stages process (know company, company needs, know self, brainstorm… | ||
| Composing algorithmic (notated) music with Python and Abjad | Tsz Kiu Pang | 32m |
Shows concrete results: Python code that performs serial transformations in one line of Python, a working Abjad pipeline that wraps LilyPond to emit engraved… | ||
| It doesn't work! | André Roberge | 25m |
friendly (and the analysis library friendly traceback) claims to turn inscrutable Python tracebacks into a three-part explanation what, where, why and offers a… | ||
| Working with penetration testers as a developer | Liam O | 33m |
Defines pen test follows a methodology and insists it needs experienced humans not scanners, separates pen tests from bug bounties and vulnerability scans, and… | ||
| A capital letter, a number, a special character, an emotional journey for the character, a subplot | Phil Nash | 30m |
Current recommendation: require at least 14 characters, accept all Unicode (spaces, emojis), and block dictionary, sequential, username‑like, or breached… | ||
| Real python, made of blocks | Katie Bell | 31m |
A block-based editor called Split Code already runs real CPython in the browser and generates editable Python while letting users drag blocks or type with… | ||
| Education Showcase | Amanda J Hogan, Nicky Ringland | 02h 00m |
Six student projects: a solar‑powered hearing aid prototypes on a Raspberry Pi Pico running CircuitPython that does FFTs, moves to a Teensy with an audio… | ||
| Setting up a machine translation service for Timor-Leste | Raphael Merx, Mel Mistica | 32m |
A volunteer project builds a machine translation service for Tetun from near-zero resources, scraping parallel articles, aligning sentences, standardizing… | ||
| Football (soccer) data analysis: A pedagogic introduction | Indranil Ghosh | 29m |
Extracts open StatsBomb data with statsbombpy and mplsoccer, draws pitches, builds directed weighted pass networks and computes centrality and clustering. Pass… | ||
| Web Security for Newcomers: how to avoid getting hacked as a software developer | Andreas Paech | 32m |
Prepared insecure Python backend and a tiny JavaScript frontend expose concrete attacks: a non-persistent XSS from echoing a query param into the page DOM via… | ||
| Collapsology: Why your biggest threat isn’t exposed RDP | errbufferoverfl | 30m |
Argument: organizational collapse follows predictable cultural failure not just unpatched services. Evidence: Neo‑Assyrian turf wars and Roman 1,700‑year… | ||
| Keynote - Ee Durbin | Ee Durbin | 52m |
PyPI’s story is concrete: uploads land in 2005 and usage explodes, a CDN rollout puts PyPI behind Fastly which serves 5.6 million requests and ~800 GB on day… | ||
| Accessibility Overlays: A Cautionary Tale | Dawn E. Collett | 35m |
Companies promise one line of code fixes that automate WCAG and ADA compliance, but overlays actually detect assistive tech (eg press alt+1), generate poor ML… | ||
| Ten Tips For Teaching Technical Topics | Tom Kunc | 29m |
Ten concrete practices and why they matter: a peanut butter sandwich exercise that forces precision, diagnosing the string plus int confusion and when to back… | ||
| Get Kookin’ With Kookaberry | Renee Noble | 31m |
A small Australian microcontroller called Kookaberry promises easier entry to MicroPython with Blockly + MicroPython options, an OLED screen and on-board… | ||
| Lightning talks | Christopher Neugebauer, Libby Berrie | 01h 26m |
Forty five-minute lightning bursts show concrete outcomes: a file-embedding codec that uses async IO and custom codecs and sometimes breaks then recovers, a… | ||
| On the Use and Misuse of Decorators | Christopher Neugebauer | 29m |
Asserts that a decorator is an expression evaluated to a function that may do anything not just wrap another function; shows concrete patterns and tools: uses… | ||
| Let’s get radical: digital privacy as a collective issue | Samantha Floreani | 30m |
Argument: privacy functions as a collective political tool to push back on surveillance, not just an individual right. Evidence: re‑identification research… | ||
| from hat import rabbit: Hacking your way through Python's import machinery for fun and profit | Joachim Jablon | 32m |
Claim: make Django believe SQL-only migrations are real Python migration modules so Django can import and run them without duplicating files. Uses the import… | ||
| Things might go wrong in a data-intensive application | Petertc Chu | 29m |
A data platform hosts several petabytes and moves several terabytes a day and yet meets new workloads it never tested: optimistic locking breaks under… | ||
| Metrics for good, not evil | Molly Rowe | 30m |
Claims that SLOs/SLIs/SLAs create a neutral, data-driven bridge between product and engineering, with SRE owning telemetry and surfacing reliability vs feature… | ||
| A Beginners Guide to GPUs for Pythonistas | Marlene Mhangami | 29m |
Shows concrete timings and tools: pandas 156 ms vs cuDF 3.56 ms for a mean, and a merge that goes from pandas 70 s vs cuDF 0.367 s, demonstrating large… | ||
| From futons to functions: the principles of good API design | Leigh Brenecki | 30m |
Connects Dieter Rams' ten principles to concrete API choices and claims five actionable priorities: iterate, make APIs learnable, contain complexity, be honest… | ||
| How PyLadies Brazil became the biggest PyLadies chapter of the world | Débora Azevedo | 34m |
Claims that PyLadies Brazil is the biggest chapter get backed by concrete work: a national network with 30 active chapters, coordination with the Python Brazil… | ||
| Doing stupid things with ASTs | Simon Carryer | 30m |
Code-level details matter: Python's ast + Inspect + Python 3.9 unparse get real ASTs, and the drawing backend uses the stdlib turtle drawing to render outputs… | ||
| Conference Open | Richard Jones | 14m |
Housekeeping, logistics and concrete numbers: the conference runs on the virtual platform Venueless, staffed by almost a hundred volunteers who spend six… | ||
| Snakes all the way down, building Worlds with Python code | Sam Bishop | 31m |
Covers concrete generators and what they actually output: Faker and Mimesis for thousands of fake profiles, a NanoGenMo project NanoGenMo (Liza Daly) that… | ||
| What was your Python up to at 3am? | Garth Kidd | 32m |
Shows how to get actionable traces from Python with OpenTelemetry and Honeycomb, wiring Flask and requests into OpenTelemetry tracing and using context… | ||
| Python Versions and Dependencies Made Easy | Sebastian Witowski | 30m |
Concrete workflow and tools: use pyenv global/local/shell to install and switch real Python interpreters rather than touching system Python, create per-project… | ||
| No, Maybe and Close Enough! Probabilistic Data Structures with Python and Redis | Simon Prickett | 26m |
Shows exact tradeoffs and runnable numbers: uses Python libraries and Redis data types to compare a set vs HyperLogLog and a Bloom Filter; adds 100,000 IDs and… | ||
| Conference Close | Richard Jones | 15m |
Announces concrete wrap‑up details: lists social events — karaoke, lightning talks hosted by Chris Neugebauer and Libby Berrie, video games, RPG one‑shots and… | ||
| nothing matches | ||