Integrail, an SFO-based agentic AI startup, has launched a no-code, build-your-own-agent platform that lets users create AI applications for any business workflow without writing code.
“It is the perfect moment to build AI that is pragmatic and can help people solve day-to-day problems,” said Anton Antich, co-founder and CEO, at the AI Field Day event in California where the company previewed Integrail Studio prior to its launch on Sept 23rd.
Agentic AI, the breakthrough after GenAI, has been on the rise recently, following an uptick in demand for self-driving AI agents that can drive workflows through autonomous actions.
Agentic AI is believed to have the potential to transform business operations with AI-based automation. A report published by the Capgemini Research Institute found that 82% of respondents plan to implement agentic AI for business workflows like email generation, coding and even data analysis, within the next three years.
Unlike chatbots that act in response to prompts and commands, agentic AI acts independently and autonomously. Armed with chaining capability that allows it to logically break down a complex job into a series of simple tasks and perform them sequentially all on its own, agentic AI has the ability to make smart decisions, perform actions and learn all in response to a single request.
A subset of intelligent AI agents constitutes an agentic AI application. Multiple agents, banded together like worker bees, perform various tasks of the same operation, leading to the fulfillment of a bigger goal.
By themselves, the agents can handle only a single task each, but by working side by side, they can pursue and solve broader problems, said Antich.
This is great for offloading repetitive tasks to programs that saves both time and cost.
The Integrail Studio is meant to be the home base for building such AI applications at scale with minimum effort.
Tom Leyden, advisor to Integrail, laid it out in a few words. “Integrail Studio allows you to create AI applications that support entire business workflows.”
“Those AI applications will use multiple agents where typical AI applications choose just one… and you can add a lot of smart stuff to it,” he added.
AI adoption in its current state is influenced by two sets of people – skeptics, who think AI hallucination makes it a hit-and-miss, and AI enthusiasts who are bullish on the technology despite its imperfections.
“The truth is in the middle, and that middle is agentic AI,” Antich said. “The ability to build smart artificial intelligence agents that are autonomous and work together with humans makes our jobs and lives much more productive and enjoyable.”
Antich gave examples of business workflows that Integrail Studio targets out of the gate. Customer support chatbot, marketing content creation, software engineering, sales forecast prediction and employee onboarding are the leading use cases, although the long-term goal, he said, is to accomplish any repetitive task, no matter the scenario.
Integrail Studio connects with a host of solutions like AI models, chatbots, custom interfaces, Discord, Chrome plugins and mobile, coming soon, as well as external business applications.
But is it truly zero-code? Antich answered with a demo. Agents on Integrail Studio can be created in a quick and fairly straightforward fashion and does not necessarily require deep AI knowledge or expertise.
Users can get started with a simple signup on the platform. To create a basic agent, they can either choose from any of the 20 featured agents from the platform’s library and customize them to specific use cases, or build them from scratch using the platform’s drag-and-drop no-code visual editor.
According to an Integrail tutorial, an agent can be built from the get-go in four simple steps. A click of a button creates a basic agent that can then be modified through the customization of parameters. Prebuilt gears available on the platform make this step fun and simple. Agents under one application can be grouped under a hierarchical order and provided a common context, memory and goal to act on.
Benchmarking is an important step to test-run agents and work out the suitability of LLMs when choosing them. Integrail Studio lets users test all AI agents against base models to evaluate and optimize their performances. Once done, the application is ready to be deployed.
“It’s a platform for very fast and very easy no-code creation of agentic applications. You can build any type of AI agent very, very fast,” said Antich.
Be sure to check out the demo from the AI Field Day event to learn how to design and deploy AI agents on Integrail Studio. To try out the platform for free, go to Integrail.ai.
