AI / ML Articles

This post provides a technical guide for developers on optimizing AI model training using Google's seventh-generation Ironwood TPU within the JAX and MaxText ecosystems.

AI & Machine Learning

TPUs

Compute

Google

11 min read

AI•2026-03-23

Build a smart financial assistant with LlamaParse and Gemini 3.1

This post explains how to build an intelligent personal finance assistant by combining LlamaParse and Gemini 3.1 Pro to extract structured data from complex financial PDFs.

OpenAI

31 min read

AI•2026-03-23

Creating with Sora Safely

This post outlines the safety-first approach taken in building Sora 2 and the Sora app as a video generation model and social creation platform.

coSTAR: How We Ship AI Agents at Databricks Fast, Without Breaking Things

This post introduces coSTAR, Databricks' methodology for reliably testing and iterating on AI agents using MLflow.

From Legacy to Lakehouse: How Mazda Accelerated GenAI for Technical Service Operations

Mazda built a GenAI-powered service assistant on Databricks Lakehouse to help hotline agents diagnose vehicle issues faster and more accurately.

What's New in Mellea 0.4.0 + Granite Libraries Release

IBM Research releases Mellea 0.4.0 and three Granite Libraries for building structured, verifiable, and safety-aware AI workflows on IBM Granite models.

Grafana

08 min read

AI•2026-03-20

Observe your AI agents: End‑to‑end tracing with OpenLIT and Grafana Cloud

This post explains how to monitor AI agents end-to-end using the OpenLIT SDK and Grafana Cloud for distributed tracing, metrics, and cost visibility.

Grafana

09 min read

AI•2026-03-20

How to monitor LLMs in production with Grafana Cloud,OpenLIT, and OpenTelemetry

This post explains how to set up end-to-end observability for LLM applications in production using Grafana Cloud, OpenLIT SDK, and OpenTelemetry.

Cloudflare

31 min read

AI•2026-03-19

Powering the agents: Workers AI now runs large models, starting with Kimi K2.5

Cloudflare's Workers AI now supports large frontier open-source models, launching with Moonshot AI's Kimi K2.5 to power agentic workloads on a unified platform.

Introducing AI Runtime: Scalable, Serverless NVIDIA GPUs on Databricks for Training and Finetuning

Databricks announces the Public Preview of AI Runtime (AIR), a serverless training stack enabling on-demand distributed GPU training on NVIDIA A10s and H100s.

Announcements

41 min read

AI•2026-03-19

Building an MCP Ecosystem at Pinterest

Pinterest describes how they built a production MCP (Model Context Protocol) ecosystem to enable AI agents to safely automate engineering tasks.

engineering-culture

infrastructure

engineering