NEAR Rabbit Hole Ep. 2: Building the Infrastructure for an Agentic AI Economy
Nothing is loaded from YouTube until you press play, and then without tracking cookies.
About this video
Episode 2 of the NEAR Rabbit Hole digs into one of the most important shifts happening in tech today: the rise of an AI-based agentic economy and how NEAR is building the infrastructure to power it.
Many chains claim they want to support AI agents. NEAR was designed for this from day zero, beginning as an AI startup before pivoting to blockchain. The result is a protocol that can scale with demand, settle transactions instantly, support millions of autonomous agents, and offer a developer experience that feels like modern software engineering instead of legacy crypto pain.
This episode covers what makes that possible: NEAR’s sharding architecture, stateless validation, fast performance, and the ecosystem’s AI-native tooling such as Shade Agents and the MCP Server. It also highlights the community projects pushing the frontier of privacy, data ownership, and user-owned AI. If you want to understand why people call NEAR the blockchain for AI, this is the place to start.
This series is sponsored by NEAR DevHub as part of the Public Goods for NEAR Protocol campaign, but all commentary reflects the creator's views.
Follow the creator: X: x.com/vinibarbosabr
Learn more and start building: Docs: docs.near.org
TIMESTAMPS: 00:07 Intro to NEAR Rabbit Hole series 00:51 An agentic economy is near 01:49 The infrastructure needed for this vision 02:29 How NEAR sharding works 03:16 A glance at performance (ep. 1) 03:48 The tech stack for an AI-agent-based economy 04:13 Community-driven AI projects and ecosystem 04:42 Shade Agents framework and MCP Server toolkit
Like, share, comment, and subscribe if you want more deep dives into NEAR’s AI, agentic, and chain-abstraction tech.
Transcript
729 wordsCaptions as YouTube publishes them, usually machine-generated, so product and people names can be misheard.
- 0:00
Near's vision is simple but bold. Become the infrastructure for an AI based agent economy. Let's talk about it. Welcome to the near rabbit hole content series that will guide you down this exciting rabbit hole that is the near protocol exploring all its facets. No worries if you can't digest it all at once. It's part of the process. Feel free to ask questions, correct me, like, share, follow, and subscribe. This series is sponsored by
- 0:26
Near Devub, part of public goods for near protocol campaign. Yet everything here reflects my genuine thoughts and opinions limited to my knowledge of these things. This is episode two and you can watch episode one on X at vinarbosabbr and on the free publication decoding.
- 0:43
You can also follow nearHub on X and visit neardeithhub.org for more information about the ecosystem. Near's vision is simple but bold. become the infrastructure for an AI based agentic economy. Many chains are trying that. But it's important to note that Near was built from the ground up with this idea in mind. First as an AI startup pivoting to a blockchain way when facing transaction challenges within the existing
- 1:15
infrastructures that were not optimized for the needed use cases. So near optimized for it. Now one of the leading projects in the AI blockchain race with infra and a text tag that compound to deliver on this vision backed by Ilia's background as a co-author of the paper attention is all you need which introduced the transformer architecture foundational to modern LLMs like chat GPT part of the optimizations happen on the merits of efficiency and scalability.
- 1:53
One person can control multiple AI agents and they can be deployed quickly for many functions. The network being able to escape according to demand while keeping its security decentralization and performance is a must. Sharding enters here splitting state into smaller fragments or shards. Sharding is the holy grail for databases and distributed systems to be able to scale to billions of users and applications. Big tech like Google, Amazon, Uber and others have
- 2:25
been using sharding for years. Every account at near belongs to some shard. All the information related to this account also belongs to the same shard. Validators can be both block and chunk producers. Blocks are made of chunks from different shards. In November 2024, near introduced stateless validation as part of nightshade 2.0. or improving efficiency, decentralization and security.
- 2:56
It fundamentally reworks the charging architecture by enabling validators to track only the shards they are assigned to significantly reducing hardware requirements and operational costs. This contributes to better decentralization, security and performance.
- 3:16
In episode one, we already mentioned how near is one of the truly decentralized and permissionless networks with the lowest fees and fashionalities. This is crucial for an agentic economy. Transactions are not only being sent fast but also fully settled irreversibly at the blink of an eye. all the while charging negligible gas fees to embrace the most vibrant use cases including micro payments.
- 3:47
Besides the infra, there is the tax tech. An agent economy needs the proper tools to operate in the best possible way. For that we have dozens of open-source tools being built with an AI based economy in mind both by the foundation for example with shape agents and the private user owned chat GPT product Ilia teased in an interview with the Nero but also by the community with projects like Nova by J Carbonel which
- 4:17
is a privacy first decentralized file sharing primitive for nears empowering user owned AI At scale we also have agent marketplaces and many other AI tools and services like bite public fraction IQ intellect and much more being built around this use case.
- 4:39
Most of these use cases are powered by Ne's shade agents framework that allows developers to build decentralized and trustless AI agents that control accounts and assets across multiple blockchains together with Near's MCP server and opensource toolkit currently with 23 tools that offers a standardized way for AI applications to integrate with Mir enabling agents to sign onchain transactions.
- 5:09
manage access keys, exchange tokens, interact with contracts, and more. Now you understand why some smart people out there are saying here is the blockchain for AI. Let me know your thoughts in the comments. Like, share, follow, and subscribe for more. See youa.