Study Says AI Search Gets It Wrong 60% of the Time

Hello!

A new study found that AI search engines get answers wrong 60% of the time, which can be a problem when they’re influencing our decisions. Meanwhile, Nvidia is betting big on physical AI, bringing machine intelligence beyond software and into robotics.

On the engineering side, Netflix continues to set the bar, processing 140 million hours of viewing data, and Microsoft is pushing an x10 speed improvement for TypeScript. If you’re interested in working with AWS Lambda, two lead engineers explain what it takes to handle billions of invocations smoothly.

Plenty of news ahead, so grab your coffee and enjoy reading Frictionless!

In the Queue

• Why Should You Delegate More

• How Zendesk Improved Job Execution

• Google Re-Starts Acquisition Talks with Wiz

• State of React Native 2024 Summary

• Leading Teams in the AI Age

Reduce Friction

AI Will Take Coders’ Jobs

Now that teams are smaller and AI-assisted coding is faster, measuring productivity by lines of code no longer makes sense. Success needs a new definition.

Delegating Complex Tasks

Letting go of responsibilities is tough, but trying to do everything yourself can make you a liability and keep your team from growing. The sooner you start offloading tasks, the less you’ll be the one holding everything back.

Executive Communication Tactics That Work For All Audiences

People won’t remember everything you say, but they’ll remember if it mattered. I know I sometimes get lost in details, and this was a great reminder to step back and focus on making my message stick.

Leading Effective Engineering Teams in the Age of GenAI

AI might not be part of the team, but integrating it requires the same leadership principles. Just like with junior devs, success depends on clear expectations, trust, and a culture where people feel safe to experiment.

Deepen Your Expertise

Handling Billions Of Invocations – Best Practices From AWS Lambda

Handling billions of Lambda invocations comes with its problems. Two lead engineers at AWS share what they’ve learned, including when to use async vs. sync execution and how to keep serverless applications running smoothly at scale.

A 10x Faster TypeScript

TypeScript is getting a major speed boost. Microsoft is building a native port to make compilation up to 10x faster, with early tests cutting VS Code’s build time from 77.8 seconds to 7.5. 

▶️ State of React Native 2024: What You Need to Know

Together with Norbert Kamieński, a React Native expert at Pagepro, we went through the State of React Native 2024. See what the future holds for our favorite mobile framework.

The Growing Significance of Observability in Cloud Native Environments

Cloud-native architectures bring new challenges, and traditional monitoring tools can’t always keep up. Observability gives teams deeper insights, helping them detect patterns, predict failures, and prevent issues before they impact users.

Increase Scalability

AI Search Engines Cite Incorrect Sources At An Alarming 60% Rate, Study Says

AI search engines are fast, but can you trust them? A study found they get answers wrong 60% of the time, bad news when they’re used for everything from coding help to business decisions.

Less is More: Improving Job Execution By Ditching The Job Executor At Zendesk

Zendesk found that ditching their traditional job executor actually made long-running jobs more reliable and efficient. Instead of relying on distributed execution, they leaned on client behavior to streamline account migrations with near-zero downtime.

NVIDIA CEO Jensen Huang’s GTC 2025 Keynote: Physical AI, Rubin Chips, and More

Jensen Huang, CEO of Nvidia, set the stage for the next wave of AI innovation at GTC 2025. The company unveiled Rubin AI chips, new GPUs, and the concept of physical AI, where AI goes beyond software into robotics and real-world applications.

How Netflix Stores 140 Million Hours of Viewing Data Per Day

Storing 140 million hours of viewing history every day is no small feat, but Netflix has it down to a science. They use Apache Cassandra for scalability, EVCache for speed, and a mix of compressed and uncompressed storage for efficiency.

Google Revives Talks To Acquire Wiz At Higher Valuation

Google is back in talks to acquire Wiz, and this time, at an even higher valuation. The cybersecurity startup has been making waves, and Google sees it as a way to strengthen its cloud security offering. If the deal goes through, it could be one of Google’s biggest acquisitions yet.

Just Cool

Astronauts Butch And Suni Begin Journey Home After Nine Months In Space

The NASA astronauts Butch Wilmore and Suni Williams are returning home. Though their initial mission was supposed to only last 8 days, a technical issue with their spacecraft extended the stay to 9 months.

Let’s Stay in Touch! 📨 

Do you have any comments about this newsletter issue or questions you want to ask? Drop me a message or book a meeting.