As of 2026, 72% of companies have embraced the regular use of artificial intelligence in at least one business function. The adoption rate of the technology has also increased, impacting nearly every industry, from technology and marketing to healthcare and retail. Yet a fundamental limitation continues to constrain its evolution, one that Christian Nonis, founder of Lumen Labs, the company behind BrainAPI, has framed his business around.
In his view, AI systems can seamlessly generate or automate, but they struggle to retain context, connect relevant facts, and build a durable understanding over time across different data sources, leading to fragmented memory recalls. BrainAPI enters this gap by positioning itself as a cognitive or "knowledge" layer for AI applications, operating primarily on the belief that intelligence without a cohesive memory is incomplete.
"We take logs, documents, transactions, chat messages, any kind of information, and build a structured network that represents that knowledge," he explains. "Then we use that to power a cognitive layer into AI agents, chatbots, search engines, and recommendation systems."
From a practical standpoint, BrainAPI is meant to sit between user data and the API system and make the data more usable by annotating, as though a human would, as well as extracting relationships, context, and meaning. In that process, it is structured to translate raw data into interconnected knowledge graphs that machines can effectively reason with, so answers are more relevant and context-aware.
Nonis highlights that the impact is significant, as 80 to 90% of the enterprise data is unstructured. This lack of structure, he insists, can limit its usability in advanced AI systems. Without a mechanism to organize and contextualize that data, he argues that even the most sophisticated models can operate with blind spots. BrainAPI addresses this by acting as an intermediary layer between data and consumption to make information accessible and, more importantly, intelligible.
"When you think of a search engine, there's often only one name that comes to mind. We are working toward that point for knowledge and memory," Nonis says, reflecting the company's ambition.
To help companies evade extensive manual engineering, BrainAPI leverages a swarm of AI agents to construct knowledge networks dynamically. Nonis explains that users can send data from existing applications through APIs, and the platform could then transform the data into an organized knowledge bank. The platform is designed to support storing and querying information from multiple source types, including documents, user behavior, and product data, so teams can uncover cryptic connections and surface insight more efficiently.
A substantial reduction in costs and deployment is another perk that BrainAPI is designed to offer. "Maintaining systems manually can be extremely costly, often requiring manual work by hundreds of engineers and data analysts. We're trying to alleviate those costs, offering on-demand deployable brains directly to the cloud," Nonis explains. With open-source capabilities, cloud-based deployments, and plugin extensibility, the infrastructure is designed to be retrofitted within existing systems.
"Developers can create a new deployment, get a URL, and start sending or querying data immediately," Nonis says. "It's designed to be simple, flexible, and fast to adopt." BrainAPI intentionally champions this modular approach with an aim to solve closed-loop ecosystems or "walled gardens."
"Large platforms often operate with vertically integrated data systems, where intelligence can be confined within proprietary boundaries, which can stifle competition, innovation, and trust," he says. "We're introducing a model that is interoperable and accessible to startups as well as scale-ups."
The company's origins reinforce this focus. Initially developed as the internal "brain" for a proactive AI assistant capable of managing emails, calendars, and workflows, Nonis highlights how the technology quickly proved more valuable as a standalone infrastructure, which resulted in a quick pivot. He reflects, "We realized the brain itself was the most powerful part, so we focused entirely on that."
With an upcoming funding round scheduled for May, BrainAPI is at a critical moment of momentum, entering a phase where visibility and credibility are as pivotal as technical execution. Nonis believes the timing aligns with broader market demand, as enterprises and developers increasingly seek scalable ways to enhance AI reasoning without exponentially increasing complexity or cost.
"We are the layer between information and whoever consumes it. Our goal is to make that information more digestible, more connected, and ultimately more useful," Nonis remarks. As artificial intelligence continues its rapid expansion, BrainAPI is building for that next wave of innovation where systems can remember, relate, and reason with engineered precision.
