Building voice-driven AI applications using LLMs
The article discusses the potential of voice-driven AI applications and the use of large language models (LLMs) in these applications. It highlights the importance of speech-to-text, text-to-speech, and the LLM itself as the three basic components for building an LLM application. The article also mentions the benefits of running application logic in the cloud, the challenges of phrase detection and endpointing, and the considerations for audio buffer management. It emphasizes the need for reliable and low-latency data flow in voice-driven LLM apps.
Original article: How to talk to an LLM (with your voice)
Featured writing
Why customer tools are organized wrong
This article reveals a fundamental flaw in how customer support tools are designed—organizing by interaction type instead of by customer—and explains why this fragmentation wastes time and obscures the full picture you need to help users effectively.
Infrastructure shapes thought
The tools you build determine what kinds of thinking become possible. On infrastructure, friction, and building deliberately for thought rather than just throughput.
Server-Side Dashboard Architecture: Why Moving Data Fetching Off the Browser Changes Everything
How choosing server-side rendering solved security, CORS, and credential management problems I didn't know I had.
Books
The Work of Being (in progress)
A book on AI, judgment, and staying human at work.
The Practice of Work (in progress)
Practical essays on how work actually gets done.
Recent writing
The intelligence briefing you're not getting
Most knowledge workers spend 45 to 90 minutes each morning manually triaging the internet. The time already exists in your day. You're just spending it on filtering instead of reading.
Dev reflection - February 11, 2026
So here's something I've been sitting with today. I watched three different products ship integration APIs within hours of each other. Same basic problem—let external systems send data in. Three co...
You were trained to suppress yourself
Organizations didn't accidentally reward the machine-self. They engineered it. And you cooperated because it worked—until now.
Notes and related thinking
Jasper is a useful tool for developing employee training.
Transform employee training with Jasper by aligning programs to business goals, engaging diverse learning styles, and using innovative methods for success.
The IMF Warns About AI's Impact on Inequality
IMF warns AI could deepen global inequality, urging policymakers to implement safety nets and retraining programs to protect vulnerable workers.
It's going to take a century for artifical intelligence to be able to perform most human jobs. But there are going to be some key developments during the next decade.
Explore how AI will transform jobs in the next decade, from enhancing security to automating coding, reshaping the future of work.