InfoWorld

Expert articles on cloud, software development, and AI

Infoworld First Look

April 10, 2024

How to test large language models

Companies investing in generative AI find that testing and quality assurance are two of the most critical areas for improvement. Here are four strategies for testing LLMs embedded in generative AI apps. Read more ▶

Image: Fully local retrieval-augmented generation, step by step

Fully local retrieval-augmented generation, step by step

How to implement a local RAG system using LangChain, SQLite-vss, Ollama, and Meta’s Llama 2 large language model.

Full-stack web development with HTMX and Bun, Part 2: Pug templating

Round out the HTMX and Bun web stack by adding Pug, a popular JavaScript templating engine that makes DOM interactions a snap.

Build or buy: Is it really even a choice?

Every software developer wants to work on challenging projects that expand their horizons. But it’s better to buy a wheel than reinvent it.

Image: Gemini Code Assist debuts at Google Cloud Next 24

Gemini Code Assist debuts at Google Cloud Next 24

Formerly Duet AI for Developers, Gemini Code Assist taps Google’s most powerful generative AI model for code completion, code generation, and code chat.

Google unveils open source projects for generative AI

Google introduced an LLM inference engine, a library of reference diffusion models, and TPU optimizations for transformer models at Google Cloud Next.

Google updates Vertex AI with new LLM capabilities, agent builder feature

Other updates include grounding applications and virtual agents in Google Search via Vertex AI and Vertex AI agent builder.

Google adds Gemini to databases to aid faster code development, migration

Gemini's availability across Google Cloud database offerings is expected to help developers code and migrate faster than Duet AI, which was integrated last year.

InfoWorld
Facebook Twitter LinkedIn
© 2024 InfoWorld
140 Kendrick Street, Building B
Needham, MA 02494