Fetch.ai Launches FetchCoder V2 to Empower Developers in Deploying Autonomous Agents

The artificial intelligence laboratory Fetch.ai Ltd. in Cambridge, UK, today announced the launch of FetchCoder V2, an AI coding assistant specifically designed for autonomous software development.

According to the company, this new AI software development assistant addresses challenges that traditional general-purpose collaborative tools cannot handle, helping developers build agents capable of independent action, learning, and interaction within decentralized systems.

The company believes an AI agent economy is taking shape: autonomous agents are moving from research labs into practical applications, becoming the norm.

Agentic AI, the fundamental process of constructing and organizing AI agents, represents a paradigm distinct from traditional software development. Agents not only need to coordinate with human developers but also with each other. When operating autonomously, agents require safeguards, as errors can accumulate rapidly. Furthermore, they operate across platforms, blockchains, and application interfaces, creating complex interactions.

This means it's impossible to inspect code execution within a single instance. Autonomous actions must be verified and validated.

"FetchCoder V2 is designed to transform ideas into intelligent software," stated CEO Humayun Sheikh. "It provides developers with the tools and confidence to turn autonomous agents into reality, eliminating guesswork and offering control at every step."

Following in the footsteps of other market tools, such as Amazon Web Services' Kiro, Fetch.ai has made FetchCoder specification-driven. Developers inform the agent what they are building before code is generated. It then validates the plan in advance, aiming to ensure consistency across the entire team. The goal is to eliminate ambiguous requirements and reduce the risk of rework.

Security is also built-in. For example, dangerous commands are automatically blocked. File modification budgets are enforced, and changes are tracked and auditable.

During the development process, testing is integrated into every step, making reliability a natural part of the workflow. If issues arise, developers can backtrack to see if the plan was flawed, identify areas needing repair, or reorganize the approach.

Under the hood, FetchCoder utilizes ASI:One, a proprietary large language model, and connects to Agentverse, Fetch.ai's platform for discovering and deploying millions of AI agents.

For users interested in connecting to the broader Web3 and blockchain technologies, FetchCoder introduces native support for the Cosmos ecosystem. This support is designed to help developers build autonomous agents that interact directly with blockchain networks.

Fetch.ai positions V2 as providing everything a general-purpose coding assistant does to help developers write code, while also aiding them in deploying autonomous agents. The company adds that this distinction is significant: building the agent economy today will shape the future.