5th annual Python Web Conf are now up on LoudSwarm for re-watch.
 |
KEYNOTE: How to Leverage Data to Combat Climate Change
by: HILLARY LOVRIC @ CLIMATE VAULT
Key points:
- Macrodata related to carbon footprint is easier to gather than microdata
- Beyond buildings and employees, a company’s carbon footprint extends to its entire supply chain
- Chesapeake Conservancy used CNNs for mapping wetlands with a 94% accuracy
- Microsoft reduced their energy consumption by 20% using AI
|
 |
KEYNOTE: Consulting for Introverted Developers
by MATT HARRISON @ METASNAKE
Key points:
- Tech consultants must put themselves out there as the quality of their work is not sufficient to attract client
- Promote yourself on LinkedIn and social media
- Don’t attend conferences: speak at conferences
- Participate in open source projects; consider writing a book
- Keep in mind: 1) Maintain an abundance attitude; 2) Invest in yourself; 3) Invest in others.
- Be helpful and trust that goodwill will come back to you
|
 |
Why you should contract test your Microservices
by SYED MUHAMMAD DAWOUD SHERAZ ALI @ ARBISOFT
Key points:
- Microservices lead to increased complexity and integration testing challenges
- Contract tests ensure that inter-application messages conform to a documented shared understanding.
- Contract consumer tests will only assert on fields it actively needs
- Consumer generated contract is verified against mock provider
- Verifies contract against provider service
|
 |
Regress to Progress: A Child’s Mindset for Growth
by RAIN LEANDER @ TEMPORAL
Key points = Use a child’s mindset:
- Be Brave: Embrace Vulnerability, Admit You Have a Fear, Do It Anyway, When You Fall Down Get Back Up!
- Play: Go For a Walk, Learn to Juggle, Go Swimming, Drum, Dance, Play a Sport…
- Explore: Listen, Ask Questions, Avoid Assumptions, Embrace Learning as Fun, Develop a Beginner’s Mind
|
 |
Data science and machine learning in wastewater intelligence
by ALEX SMOLYAK @ KANDO
Key points:
- ML algorithms coupled with lower cost sensors can model water quality at a much higher granularity than is done with traditional sensors and telemetry.
- Kando is striving to collect public and proprietary data, deploy their lower-cost sensors more liberally across the areas of interest, train models on all of that data, and generate actionable insights.
|
 |
TUTORIAL: Refactoring without a net
by GIL ZILBERFELD @ TESTINGIL
Key points:
- Refactoring is risky if you don’t have tests and/or you don’t really trust your code;
- When refactoring, don’t “fix” or improve anything;
- A good tool to create a basic safety net is ApprovalTests;
- Eventually replace ApprovalTests with real tests.
|
 |
How Visual AI Makes Testing Easy
by ANDREW KNIGHT @ APPLITOOLS
Key points:
- It is difficult to spot visual differences;
- Visual AI creates a screen capture the first time it runs and compares it against the previous screen capture. Changes have to be manually reviewed and confirmed;
- Visual AI accelerates cross-platform testing.
|
 |
Minimum Viable Migrations
by BEN ELLERBY @ ALEIOS
Key points:
- MVM is a usable migrated version of an application from one paradigm to another with a subset of features
- MVM allows to test hypotheses early, learn faster, reduce waste, ship to customers quicker, validate assumptions, etc.
- You need a strategy to run 2+ systems in parallel transparently to users and an exit strategy from the old system. Consider moving low priority features last.
- Use event driven architectures
|
 |
Level Up Your Developer Experience
by ROCHÉ COMPAAN @ SIX FEET UP
Key points:
- Use Kubernetes as a development environment
- You are the architect, Kubernetes is the builder. You write the manifests, Kubernetes takes care of the rest, and stays there for maintenance
- There is no vendor lock-in and you only need to know a few basic concepts for development
- Roché outlined steps for setting up a Kubernetes cluster and suggested development tools
|
 |
Building an IoT App with InfluxDB, Python, Flask And Plotly
by ANAIS DOTIS-GEORGIOU @ INFLUXDATA
Key points:
- InfluxDB is a highly scalable time series database.
- It is moving to IOx as its underlying storage foundations allowing more SQL capabilities.
- Presentation ends discussing collecting environmental sensor data using an Arduino.
- The data is collected into InfluxDB and summarized and graphed with plotly and Flask.
|
