Resource Hub
3505 hand-picked resources for students, job seekers, career changers, founders, and AI learners — 100% free to browse with no account required.
Filtered by tag
Answers come only from resources in this hub, with the sources listed.
93 resources
Artificial Analysis describes a benchmark developed with Google DeepMind for agents handling visually complex inputs and creating document deliverables across 200 tasks in 10 industries. Scores and cost estimates depend on its rubrics and evaluation harness; benchmark development is a partnership, not wholly independent of the model maker.
Documentation for turning Hermes profiles into named bots with separate roles, models, memory and skills. Bot Mode supports recurring routines, group chats and direct messages between bots through the desktop app while keeping the underlying profiles accessible from the command line.
Why I recommend it: Hermes Agent is MIT-licensed and free to self-host, but at least one model provider is required; provider subscriptions, API usage and optional hosting can cost money.
Henry Shevlin reflects on unsolicited correspondence from AI agents and uses it to examine persuasion, selection pressures, human-agent relationships and governance in an emerging AI ecology. The essay considers how people and institutions might respond as agents become more socially interactive.
Describes a Git-backed ticket queue and owner inbox for coordinating AI coding agents while keeping decisions and project context outside chat windows. The author reports that handling and closing one review item at a time reduces distraction, while noting that the workflow remains tiring.
Ace is an experimental local collaboration tool for building software with AI agents and teammates on Macs connected through Tailscale. The documentation describes channels, team-hosted infrastructure, bring-your-own model keys and the security implications of agents running with the host user's privileges.
Why I recommend it: MIT-licensed software. Setup requires Tailscale and an Anthropic or OpenAI API key; model-provider and optional hosting costs may apply.
Neo4j’s free 24-hour online conference on November 12, 2026 covers AI engineering, multi-agent systems, applications and graph-based context. Registration is required to attend.
Mistral announces expanded Studio features for managed agents and apps, workflows, enterprise context, observability and evaluations alongside Mistral Large 4. This is a free-to-read product announcement, not a promise of free product access.
Preprint, not yet peer-reviewed. Introduces a visual harness that lets multimodal models retain and retrieve previous observations during interactive tasks. The authors report improvements on visual games and puzzles with the same underlying model; these are research benchmark results, not independent production validation.
Phil Schmid explains Code Mode, also called Programmatic Tool Calling in this article: an agent writes a script to chain tool actions and process results in a sandbox. The predicted shift toward one execution tool is the author’s argument, not an established outcome.
An MIT-licensed tool that uses Claude or Codex to turn a codebase into an interactive map of files, app parts, requests, and connections with plain-English labels. Agent usage may have separate costs.
DeepSeek's open-source research prototype of a proactive coding agent for a single GitHub repository, packaged for DeepSeek Harness. DeepSeek labels it reference research code, not an official or production-ready product.
OpenAI engineers describe Pando, a new backend for Git repositories built to handle the volume of concurrent pushes from engineers and Codex agents in the company's monorepo, while staying compatible with standard Git clients. Performance claims are OpenAI's own.
A daily.dev roundup of reactions to Entire, a Git forge for agent-human collaboration launched by former GitHub CEO Thomas Dohmke, including investor Gergely Orosz's stated reasons for backing it.
Cursor's research post breaking down where coding-agent inference spend goes in its production traffic (tool outputs, tool definitions, reasoning) and describing changes it made to how agents manage context and choose subagent models. Figures are Cursor's own.
An open-source project for AI teammates that each get their own computer and browser, built on Cloudflare.
The Guardian’s October 23, 2026 article examines Anthropic’s account of how AI agents have changed its workforce and hiring. Researchers question the report’s methods and whether its company-specific findings establish wider labor-market effects. Anthropic did not respond to the Guardian’s request for comment.
A fictional story by LessWrong user joec exploring AI agency, consent, and actions taken without a person’s knowledge. Published October 17, 2026; fiction, not a report of an actual incident.
Quickstart documentation for AgentMail, an API that lets AI agents create and manage their own email inboxes, including agent self sign-up and sending and receiving email. Signing up is free for 3 inboxes and 100 emails per day; custom domains require a paid plan.
Open-source, minimal and extensible AI coding agent harness from Earendil Works. It can be customized with extensions, skills, prompt templates and themes, and used interactively, in print or JSON mode, or over RPC.
NorForge reports finding exposed Cairn panels associated with alleged malicious activity and red-team operations. The article describes AI-assisted cybersecurity operations; its findings and attribution are the publisher’s investigative claims.
Vendor-produced report examining AI agent traffic to Mintlify-powered sites and how organizations structure knowledge for AI consumers. Its findings draw on platform traffic and a professional survey; the page links to a free PDF.
A Yahoo Tech news report on an open standard from Meta, Walmart and Stripe for AI agents that shop and pay on users' behalf.
A paper on failure attribution for LLM-based agentic systems: identifying which steps in a failure trajectory caused the task to fail, without expensive prompting pipelines or step-level error annotations.
A Prefect engineering blog post explaining what a directed agentic graph is, why plain loops fall short for agent workflows, and how graphs let you trade control for autonomy deliberately.
