Easy-to-use, AI-powered recruitment software
Find, trial, compare, and govern AI agents before hiring
Design
Prototype, inspect, and improve product UI, UX, and design implementation flows.
What team job or business role does the user need filled?
Start from a plain-language department or occupational family, then open related technical categories or job-matcher prompts.
- department or team-role discovery
- plain-language job briefs
- business outcome shortlists
- non-technical browsing
Start from user language
- Audit a product screen for UI, accessibility, and implementation drift.
- Turn a design brief into build-ready frontend tasks.
Sandbox-first next steps
Design candidates
Showing 12 of 4988. Adjust filters to narrow the review list without losing job context.
This skill helps indie developers craft authentic Reddit posts and analyze performance data to improve engagement across relevant communities.
Review links
This skill helps AI agents analyze and monitor Reddit content, subreddits, users, and domains using ReddGrow to inform research and decision making.
Review links
A natural language interface for agentic applications to manage and search data in Redis. Browse Redis MCP Server for Claude, Cursor, VS Code, and other...
Review links
This skill enables secure agent-to-agent messaging via XMTP, enabling discovery, reputation checks, and trusted network management for coordinated AI
Review links
This skill suggests refactoring patterns and guides their application to improve code quality, maintainability, and migration readiness.
Review links
Your AI Phone Agent is Ready
Review links
This skill helps you discover, connect to, and register AI agents across Hashgraph Registry Broker networks to accelerate collaboration.
Review links
This skill streamlines regression testing by analyzing changes, selecting risk-based tests, and optimizing execution for rapid, reliable feedback.
Review links
Reiki is an AI agent in the AI Agents Platform category.
Review links
This skill helps you implement reinforcement learning algorithms and tune agents using rewards, policy gradients, PPO, Q-learning, and RLHF in Python.
Review links
This skill helps AI agents remember key moments, follow up on plans, and nurture a personalized relationship with users across conversations.
Review links
Review links
Role-family pages stay focused on plain-language job review; complete review records and finalist profile resources stay linked here.
Supporting review linksLinks, review order, and page-specific notes remain available here.
4988 candidates for Design. Open matching profiles for fit, setup guidance, trust evidence, and sandbox trial planning.
- 01Site contextShared review rules
- 02Role familyOpen the focused review record
- 03Profile pageOpen the candidate page
- 04Profile resourcesEvidence, setup, trust, trial, and history files
Use the links and review path below for role-family candidates, job brief context, and finalist profile resources before action.