Hacker News with Generative AI: AI Agents

Ask HN: How to teach agentic AI? Please share your experience (ycombinator.com)
I started teaching agentic AI at our cooperative (Berlin). It is a one day intense workshop where I: <p>1. Introduce IntelliJ IDEA IDE and tools 2. Showcase my Unix-omnipotent educational open source AI agent called Claudine (which can basically do what Claude Code can do, but I already provided it in October 2024) 3. Go through glossary of AI-related terms 4. Explore demo code snippets gradually introducing more and more abstract concepts 5.
Show HN: Hyperbrowser MCP Server – Connect AI agents to the web through browsers (github.com/hyperbrowserai)
This is Hyperbrowser's Model Context Protocol (MCP) Server. It provides various tools to scrape, extract structured data, and crawl webpages. It also provides easy access to general purpose browser agents like OpenAI's CUA, Anthropic's Claude Computer Use, and Browser Use.
Arcade raises $12M from Perplexity co-founder's fund to make AI agents less bad (techcrunch.com)
Arcade, an AI agent infrastructure startup founded by former Okta exec Alex Salazar and former Redis engineer Sam Partee, has raised $12 million from Laude Ventures.
Show HN: Computer – Build Your Manus AI Agent with an OSS macOS Sandbox (github.com/trycua)
Create and run high-performance macOS and Linux VMs on Apple Silicon, with built-in support for AI agents.
Launching the Naptha Stack v1 Beta (naptha.ai)
We're thrilled to announce the launch of the Naptha AI Stack v1 Beta, our groundbreaking open-source platform for autonomous AI agents that enables unprecedented collaboration across different AI agent architectures.
How we improved GPT-4o multi-step function calling success rate by 4x (xpander.ai)
AI Agents will usher in a new era of human-computer interfaces, automation, personal AI assistants and AI employees. Function calling enables AI Agents to execute complex, multi-step workflows with precision, and plays a big part in fulfilling the promise of AI Agents.
Agent Graph System boosts GPT-4o multi-step function calling success rate by 4x (xpander.ai)
AI Agents will usher in a new era of human-computer interfaces, automation, personal AI assistants and AI employees. Function calling enables AI Agents to execute complex, multi-step workflows with precision, and plays a big part in fulfilling the promise of AI Agents.
SuperPrompt (github.com/NeoVertex1)
This is a project that I decided to open source because I think it might help others understand AI agents. This prompt took me many months and is still in phase of forever beta. You will want to use this prompt with Claude (as custom instructions in the project knowledge) but it also work with other llms.
Best Practices for Prompt Writing – DigitalOcean Documentation (digitalocean.com)
DigitalOcean GenAI Platform is an offering for building AI agents on GPU-powered infrastructure. Use the platform to create AI applications with foundation models and agent routes, knowledge bases, and Retrieval-Augmented Generation (RAG) pipelines. See Features for more information.
A review of OpenAI o1 and how we evaluate coding agents (cognition.ai)
Devin is an AI software engineering agent that autonomously completes coding tasks. We’ve been testing OpenAI’s new o1-mini and o1-preview models with Devin for the past several weeks and are excited to share some early results. To contextualize these results we will also discuss our evaluation methodology and our technical approach to building reliable coding agents.
Show HN: Outlit – AI Agents to Empower Non-Legal (outlit.ai)
AI Agents That Matter (aisnakeoil.com)
SceneCraft: An LLM Agent for Synthesizing 3D Scenes as Blender Code (openreview.net)
Show HN: Use functional tokens for AI agents to simplify app workflows (nexa4ai.com)