Brevian: The No-Code Enterprise Platform for Effortlessly Creating AI Agents

Sunnyvale-based Brevian is revolutionizing the way businesses create custom AI agents. Initially, the company is concentrating on support teams and security analysts—two fields with well-defined use cases and training datasets—but it has plans for future expansion into other sectors. Today, Brevian emerges from stealth mode and announces a significant $9 million seed funding round.

Formally styled as BREV/ΛN, Brevian was co-founded by Vinay Wagh (CEO) and Ram Swaminathan (CTO), who bring diverse experiences to their startup journey. Wagh previously served as product director at Databricks and led product initiatives at Bracket Computing. While at Databricks, he focused on preparing the company for enterprise sales and launching the Databricks SQL Serverless product in February. "When I observed technologies like ChatGPT, I wondered how they could be applied in the enterprise," Wagh explained. "It became clear that addressing the gaps in this space was essential for realizing AI's potential." His partnership with Ram Swaminathan began through a mutual friend.

While Wagh's expertise lies in data, Swaminathan brings a wealth of knowledge from computer science and machine learning. After his academic tenure, he spent nine years at Bell Labs and 17 years at HP Labs, contributing to projects ranging from coding theory to cryptography and early machine learning initiatives. He joined LinkedIn as the leader of the AI trust team before co-founding Brevian as CTO in late 2022.

The duo emphasizes that security remains at the heart of all enterprise generative AI efforts. Companies must address personal information leaks to gain confidence in AI deployment. In the early days, many enterprises opted to ban ChatGPT due to concerns about sensitive data being unintentionally exposed within chat prompts.

Brevian initially prioritized security measures. "We quickly developed models capable of detecting personally identifiable information (PII)," Wagh stated. "As we observed an uptick in prompt injection attacks, we established an intent-based system to detect these intrusions. Our approach involved gathering all prompts that confuse AI alignment and running a model to catch them."

Interestingly, while enterprises remain apprehensive about security in large language models, they often struggle to pinpoint their exact concerns beyond data leaks. The real obstacle, as the Brevian team discovered, was not just security but the difficulty in creating systems that directly address enterprise needs.

“Our vision centers on empowering business users to leverage AI for simplifying their daily tasks,” Wagh shared. “This aligns with our goal of expanding from security to creating versatile AI agents.”

Brevian's seed funding round was led by Felicis partner Jake Storm, marking the firm's largest seed investment to date. Storm observed a shift in focus over the past year: “In 2023, discussions revolved around AI infrastructure, but the spotlight is now shifting to applications—particularly the importance of security.” He confidently predicts that “2024 will herald the year of AI applications, and we believe Brevian is a frontrunner in this realm."

With this funding, Brevian aims to accelerate product development and grow its team to address increasing customer demand as part of its early release program.