#14: eBay, Airbnb, and Y Combinator Just Went All-In on AI Agents

AgentPulse: Your Weekly Dose of AI Agents News.

Author

Oliver Parker
May 12, 2025

In partnership with

Welcome back, AI Agent Enthusiast!

In today’s Agent Pulse:

  • 📢 The top AI Agent news you shouldn’t miss

  • 🚀 Top AI agents Leaderboard

  • ✨ Featured AI Agents

  • 🎓 Free courses to level up your AI agent game

  • Agents Protocols.

Submit your AI Agent to the leading AI Agents Directory & Marketplace.

Request Custom AI Agent from trusted AI Agent builders

📢 Top Stories This Week

eBay rolled out its first consumer-facing AI shopping agent, delivering real-time, hyper-personalized product recommendations and guidance based on each shopper’s. Seamlessly integrated into the browsing experience, the agent can suggest gifts or assemble outfits interactively. This marks a key step in eBay’s “AI-first” strategy, aiming to make product discovery effortless and boost engagement by leveraging agentic AI on its marketplace. (source)

Airbnb deployed AI customer-service agent to about half of U.S. and reported that the agent, which handles common travel-support queries in the app, has already reduced contacts to human support agents by roughly 15%. The agent is expected to become more personalized and “agentic” over time, allowing Airbnb to resolve bookings and travel issues faster while freeing human staff for complex cases. (source)

TensorStax raised a $5 million seed round (led by Glasswing Ventures) to develop deterministic AI agents for data. Its platform plugs into existing data stacks (Airflow, dbt, Snowflake, etc.) and uses LLMs to automate complex pipeline tasks while maintaining strict schema and reproducibility requirements. By handling routine data transformations accurately, TensorStax’s agents aim to dramatically speed up the “rigid” data engineering process and let teams focus on higher-level modeling work.

StackAI, a San Francisco no-code startup (YC W23), announced a $16 million Series A led by Lobby Capital. The funding will expand its platform that enables non-technical users to build custom AI agents via drag-and-drop workflows. StackAI’s agents can integrate language models with enterprise tools (e.g. Snowflake, Salesforce) to automate tasks like data entry or content aggregation, and the company plans to launch a marketplace of pre-built agent templates to accelerate deployment

Signals unveiled a cloud-based “AI Employee” platform to automate front-line tasks across sales, marketing, customer service and HR. Its agents come pretrained on over a billion sales conversations and can deploy instantly, integrating with a company’s systems (CRM/ERP) to handle common tasks like scheduling appointments or resolving basic service cases. By providing these ready-to-run conversational agents, Signals aims to let businesses engage customers across phone, email, chat and social media without adding headcount.

Umanitek, Swiss AI startup Umanitek AG launched Umanitek Guardian, its first AI agent product, focused on online safety. Built on a decentralized knowledge graph, the agent lets organizations (like law enforcement) upload and query millions of verified content “fingerprints” (image/video hashes) in natural language. This helps partners collaboratively detect and track harmful or illegal content across platforms without exposing raw data, addressing internet safety and privacy challenges in the AI era.

Alation launched its Data Products Builder Agent, an AI tool that turns raw enterprise data into governed, AI-ready data products. The agent works via user prompts to identify relevant data, auto-generate data-product specs, enforce governance standards, and publish datasets into Alation’s data marketplace. This automation removes tedious curation work for data teams and ensures that AI and analytics projects have clean, documented data assets at scale.

Y Combinator has released its Summer 2025 “Request for Startups” declaring 2025 the “year of AI agents.” The accelerator is particularly interested in full-stack AI startups—companies that not only develop AI agents but also operate entire businesses powered by them, such as AI-driven law firms or clinics. This call aims to encourage founders to build AI-native companies that integrate agents as core operational components, rather than as mere features. Apply by May 13

Explore 1300+ AI Agents on Dynamic Landscape Map

Modernize your marketing with AdQuick

AdQuick unlocks the benefits of Out Of Home (OOH) advertising in a way no one else has. Approaching the problem with eyes to performance, created for marketers with the engineering excellence you’ve come to expect for the internet.

Marketers agree OOH is one of the best ways for building brand awareness, reaching new customers, and reinforcing your brand message. It’s just been difficult to scale. But with AdQuick, you can easily plan, deploy and measure campaigns just as easily as digital ads, making them a no-brainer to add to your team’s toolbox.

🚀 TOP Agents Leaderboard

View Leaders:
🥇 ACMEBOT: Automated content creation AI agent for businesses.
🥈 TheLibrarian.io: WhatsApp AI Personal Assistant
🥉 Agentverse: Search and Discover AI Agents

Upvote Leaders:
🥇 ACMEBOT: Automated content creation AI agent for businesses.
🥈 TheLibrarian.io: WhatsApp AI Personal Assistant workflows
🥉 Vessium: Platform for building agentic workflows

Explore Dynamic AI Agents Leaderboard

✨ Featured Agents

  • TheLibrarian.io: WhatsApp AI Personal Assistant

  • TeammatesAI: Autonomous AI Teammates

  • Agentverse: Search and Discover AI Agents

  • DoozerAI: AI-driven digital workers that transform business operations

  • Twig: Automate Tier 1 support with AI Agents

  • TensorStax: Autonomous AI Agents for Data Engineering

🎓 FREE Courses

Agents Protocols

  • MCP (Model Context Protocol): Standard way for tools to share context and data with AI models in both directions, so agents get what they need in real time. Introduced by Anthropic.

  • A2A (Agent‑to‑Agent Communication Protocol): Common message format that lets agents talk, discover each other’s abilities, and coordinate tasks securely. Introduced by Google.

  • ACP (Agent Communication Protocol): A simple, consistent structure for agents to call and control each other, making multi‑agent workflows easier. Introduced by IBM under BeeAI project.

  • ANP (Agent Network Protocol): A peer‑to‑peer framework for agents to verify identities and connect directly—think HTTP, but for AI agents. Introduced by Gaowei Chang

  • AGORA:  A flexible meta‑protocol that picks the best way (structured calls, natural language, or code) to let large networks of agents collaborate without a central hub. Introduced by researchers at the University of Oxford and Eigent AI.

SPONSOR US

Get your product in front of over 8.5k+ readers

Our newsletter is read by VCs, founders, engineers, managers and tech professionals.

FEEDBACK

Did you like the newsletter?

Login or Subscribe to participate in polls.

Reply to this email, I would LOVE to hear from you!

I appreciate your support.

OP & Team