Contact Us
‍
24/7
‍
Python BlogDjango BlogSearch for Kubernetes Big DataSearch for Kubernetes AWS BlogCloud Services
<< ALL NEWS

4 Ways to Reduce Friction in Software Development

September 10, 2026
|
4 Ways to Reduce Friction in Software Development

Building software isn't just about writing code. Time gets lost clarifying requirements, reproducing results, managing infrastructure, and getting applications into production.

At an IndyPy Lightning Talks meetup, four speakers showed how to remove that friction: validate ideas earlier, make notebooks reproducible, simplify deployment, and manage infrastructure like application code.

1. Validate Ideas Before Writing Code

David Maish, Six Feet Up Service Delivery Lead | Figma Make

AI-powered prototyping can help teams catch misunderstandings before they become expensive development work.

David demonstrated how Figma Make turns prompts into interactive, shareable prototypes. Instead of walking stakeholders through static mockups, teams can send a clickable prototype ahead of a meeting and use the time to discuss what works, what doesn't, and what needs to change.

In one example, David spent roughly two hours reviewing a prospective client's spreadsheet workflow and building a clickable prototype to show what a web application could look like.

The takeaway: Give stakeholders something to interact with early. Better feedback before development means fewer assumptions and less rework later.

Watch David’s Lightning Talk, “Using Figma Make to Create Interactive Prototypes”:

2. Make Python Notebooks Reproducible

Glenn Franxman, Six Feet Up Senior Architect | marimo

Traditional Jupyter notebooks can introduce hidden dependencies and inconsistent results when cells execute out of order.

Glenn introduced marimo, a reactive notebook environment that automatically tracks cell dependencies and reruns affected code. Unlike Jupyter's JSON-based notebooks, marimo stores notebooks as plain Python files, making them easier to manage in version control, review, and automate.

He also demonstrated interactive SQL queries, visualizations, and the ability to run notebooks directly from the command line.

The takeaway: Treat notebooks as maintainable code, not disposable experiments. Reproducible workflows are easier to share, automate, and move into production.

Watch Glenn’s Lightning Talk, "marimo Notebooks":

3. Simplify the Path to Production

Calvin Hendryx-Parker, Six Feet Up CTO and AWS Hero | FastAPI Cloud

Deploying an API can take more effort than building it.

Calvin demonstrated how FastAPI Cloud takes a Python API from a local project to a live application in just a few commands, without manually configuring Dockerfiles, YAML, or cloud infrastructure.

The platform also provides live logs, environment variable management, database integrations, and CI/CD support, reducing the infrastructure developers need to configure themselves.

The takeaway: Deployment shouldn't become a separate engineering project. Standardize the path to production so developers spend less time configuring infrastructure and more time building features.

Watch Calvin’s Lightning Talk, “FastAPI Cloud”:

4. Manage Cloud Infrastructure Like Application Code

Rafael Rodriguez, AWS Developer | AWS CDK

Manual cloud configuration creates drift. Months later, teams may struggle to understand how resources were deployed, what depends on them, or whether they're safe to remove.

Rafael compared AWS CloudFormation, the AWS Serverless Application Model (SAM), and the AWS Cloud Development Kit (CDK), showing how CDK lets developers define infrastructure using familiar Python concepts.

He also highlighted how reusable CDK constructs can help platform teams establish standards that developers inherit, rather than repeatedly implementing infrastructure requirements.

The takeaway: Infrastructure as Code isn't just about automating deployments. It's about making cloud environments easier to understand, test, maintain, and govern as teams scale. Start with one reusable construct for the resource your teams rebuild most often, and let every new project inherit it.

Watch Rafael’s Lightning Talk, “AWS Infrastructure as Code with Python”:

Keep Learning with IndyPy

From validating requirements earlier to making infrastructure more repeatable, these talks reinforced a common lesson: removing friction at every stage of development helps teams spend less time on rework and more time delivering value.

Join IndyPy for future meetups, technical presentations, and conversations with the Python community.

How can we assist you?
Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.