Learn - Share - Innovate - Elevate

DataOps Newsletter Volume 17 - May 2025

DataOps Labs

May 15, 2025

Join Amazing In Person Sessions

1) Be Part of AWS Community Day Bengaluru -

https://acd.awsugblr.in/

Workshops were Announced. Pick anyone of your choice - Register

https://acd.awsugblr.in/ to be get benefit out of workshop

Online Events:

Deploying a Flask Application to AWS Fargate with Amazon Q CLI

Topic: Building a Full-Stack Budget Tracker Application with Amazon Q CLI

Innovate

What is Amazon Nova-ACT?

Amazon Nova-ACT is an AI-driven web automation tool that lets users describe tasks in natural language. It translates those instructions into precise browser actions and executes them on live websites. Instead of writing code like driver.find_element_by_xpath(...), you can say “log in and show me my recent orders.”

This turns web automation from a developer’s headache into a user-friendly, AI-powered assistant.

How It Works

Nova-ACT runs in your Python environment and leverages a local or headless browser session. It uses the following process:

Interpret natural language commands
Translate them into actionable steps
Execute those steps in a browser
Observe results and iterate automatically
Complete the task end-to-end

Behind the scenes, it’s using AI to bridge the gap between what you mean and what needs to happen in the browser.

Usecase worked for me : Retrieving Uber Trip Receipts

Let’s say you want to download the receipt for your most recent Uber ride. Doing this manually means logging into Uber, navigating to trip history, scrolling through, and clicking "View Receipt." It’s mundane and repetitive.

import fire  # type: ignore

from nova_act import NovaAct


def main(user_data_dir: str, headless: bool = False) -> None:
    with NovaAct(
        starting_page="https://riders.uber.com/trips",
        user_data_dir=user_data_dir,   # <-- Correct: Use the parameter passed!
        headless=headless,
        clone_user_data_dir=True,
    ) as nova:
        nova.act("I am already logged into my Uber account, proceed directly to trip history")
        nova.act("identify the most recent trip charged in US Dollars")
        nova.act("View the receipt for trip")


if __name__ == "__main__":
    fire.Fire(main)

With Nova-ACT, this process is automated using a few natural language instructions:

This script:

Uses your existing Chrome profile (so you're already logged in)
Opens the Uber trip history page
Locates your most recent trip charged in USD
Opens the receipt page automatically

Stay tuned on my blog for end-to-end working solution with demo -

https://blog.dataopslabs.com/

Elevate

Feel free to have a glance at blogs and share your thoughts