SignalWire AI Agents and the No-Code (R)evolution

Build a virtual assistant without any engineering experience

Product Manager

Daniele Di Sarli

What if you wanted to replace your old-fashioned IVR with something more natural, something more human-like, while maintaining the cost and scalability advantages of a fully automated solution?

Not long ago, the most likely path would have been to use your engineering resources to integrate your product with a third-party solution - often with clunky results. We believe that you shouldn’t waste your engineering resources reinventing the wheel. That’s why we recently introduced SignalWire AI Agents - virtual assistants you don’t need engineering expertise to build, just basic technology skills.

SignalWire AI Agents are defined using SignalWire Markup Language (SWML), which only requires basic JSON or YAML skills to use. Whenever a change is needed you don’t have to go through the engineering cycle of writing the code, committing, deploying - you can just change a configuration file. This is especially useful for conversational AIs like SignalWire AI Agents, since you may want to tune your prompts in fast iterations.

A user interface for AI Agents

Editing a YAML document is great with respect to writing complex integration code. Still, it requires a basic level of familiarity with computer languages, which some of your team members may not have.

The person fine-tuning the prompts doesn’t need to be an engineer with our brand new user interface for AI Agents. Every member of your team will now be able to log into their SignalWire space, navigate to AI Agents, and create a new agent.

AI Agents are also now integrated into Call Flow Builder, so that anyone, with or without coding experience, can define their call flows which include conversational AI.

Not just a user interface

AI Agents are now first-class resources of the SignalWire platform, just like SIP endpoints, SWML scripts, and video rooms. This means, for example, that if you own a phone number, you can decide to handle inbound calls to that phone number with an AI agent. This makes AI Agents even easier to use.

Integrating advanced AI solutions often requires significant engineering resources and clunky third-party solutions, but SignalWire AI Agents, powered by SWML and a user-friendly interface, have changed the game. With basic technology skills and plain text language, users can make quick and effortless changes to their virtual assistants, eliminating the need for time-consuming coding and deployment cycles.

The evolution of AI has never been more user-centric, making AI adoption accessible to all, regardless of technical background.

If you have any questions as you build your new AI Agent, drop by our Community Slack or Discord to ask our team of experts!

Related Articles