Cambridge, U.K.-based artificial intelligence lab Fetch.ai Ltd. today announced the launch of FetchCoder V2, an AI coding assistant purpose-built for autonomous software development.
According to the company, its new AI software development assistant addresses challenges that traditional, general-purpose copilots can’t by helping developers build agents that can act, learn and interact independently across decentralized systems.
The company argued the AI agent economy is coming into its own: Autonomous agents are becoming the norm, moving out of research labs and into real production.
Agentic AI, the fundamental process of building and organizing AI agents, represents a different paradigm from traditional software development. Agents must coordinate not just with human developers, but with one another. Agents also need guardrails when operating autonomously, because mistakes can compound quickly. And they operate across platforms, blockchains and application programming interfaces simultaneously, creating complex interactions.
All that means that it’s not possible to check code runs in single instances. Autonomous behavior must be verified and validated.
“FetchCoder V2 is about turning ideas into intelligent software that works,” said Chief Executive Humayun Sheikh. “It gives developers the tools and confidence to bring autonomous agents to life, without guesswork and with control over every step.”
Following in the footsteps of other tools on the market, such as Amazon Web Services Inc.’s Kiro, Fetch.ai made FetchCoder spec-driven. Developers tell the agent what they’re building before it generates a single line of code. It then validates the plan up front, aiming to ensure alignment across the entire team. The goal is to eliminate ambiguous requirements and reduce the risk of rework.
Safety is also built in. For example, dangerous commands are automatically blocked. File modification budgets are enforced and changes are tracked and auditable.
During development, tests are baked into every step, making reliability a natural part of the workflow. If something goes wrong, developers can backtrack through progress to see whether the plan is flawed, identify what needs to be fixed, or reorganize the approach.
Under the hood, FetchCoder uses ASI:One, a proprietary large language model and hooks into Agentverse, Fetch.ai’s platform for discovering and deploying millions of AI agents.
For users interested in connecting to the broader landscape of Web3 and blockchain technology, FetchCoder introduces native support for the Cosmos ecosystem. That support is intended to help developers build autonomous agents that interact directly with blockchain networks.
Fetch.ai positioned V2 as providing everything general-purpose coding assistants do to help developers write code, while also helping them ship autonomous agents. The company added that the distinction matters: building the agent economy today will shape the future.
Image: News/Microsoft Designer
Support our mission to keep content open and free by engaging with theCUBE community. Join theCUBE’s Alumni Trust Network, where technology leaders connect, share intelligence and create opportunities.
- 15M+ viewers of theCUBE videos, powering conversations across AI, cloud, cybersecurity and more
- 11.4k+ theCUBE alumni — Connect with more than 11,400 tech and business leaders shaping the future through a unique trusted-based network.
About News Media
Founded by tech visionaries John Furrier and Dave Vellante, News Media has built a dynamic ecosystem of industry-leading digital media brands that reach 15+ million elite tech professionals. Our new proprietary theCUBE AI Video Cloud is breaking ground in audience interaction, leveraging theCUBEai.com neural network to help technology companies make data-driven decisions and stay at the forefront of industry conversations.
