WisdomAI Raises $50M Led by Kleiner and Nvidia
▼ Summary
– WisdomAI raised $50 million in Series A funding led by Kleiner Perkins with participation from NVentures, following a $23 million seed round six months prior.
– The startup provides AI-driven data analytics that answer business questions using natural language queries on structured, unstructured, and “dirty” data.
– It avoids LLM hallucinations by using LLMs only to write queries for data retrieval, not to generate answers, ensuring responses are based on actual data.
– WisdomAI has grown from two to around 40 enterprise customers since its late 2024 launch, including companies like Cisco and Patreon, with rapid usage expansion.
– The company introduced an agentic feature that alerts users in real time to important changes in monitored data, making analytics proactive rather than static.
WisdomAI, a promising new player in the artificial intelligence data analytics field founded by Rubrik co-founder Soham Mazumdar, has successfully secured $50 million in Series A funding. This significant investment round was spearheaded by Kleiner Perkins and saw substantial participation from NVentures, the venture capital division of technology giant Nvidia. This financial milestone arrives just half a year after the company’s initial $23 million seed funding, which was led by Coatue.
The platform developed by WisdomAI delivers AI-powered data analytics capable of interpreting business inquiries using structured, unstructured, and even imperfect or “dirty” data, information that hasn’t been corrected for typographical errors or inaccuracies. Business professionals can pose questions using everyday language, such as inquiring about the number of potential customers in their sales pipeline or identifying obstacles to closing deals within the current quarter.
A standout aspect of WisdomAI’s methodology involves its innovative solution to the common issue of large language model (LLM) hallucinations. Instead of employing LLMs to generate direct answers, the system uses them exclusively to formulate database queries. This means if an LLM produces an inaccurate or nonsensical query, the result is merely an ineffective data retrieval attempt rather than a fabricated response. This design significantly reduces the risk of misleading information.
Underpinning this approach is what the company terms its “enterprise context layer,” a proprietary logic system that analyzes and comprehends customer data. The entire founding team, including Mazumdar, previously collaborated at data security firm Rubrik, providing them with extensive expertise in enterprise data storage solutions. Mazumdar departed from Rubrik in 2023.
Since its official launch toward the end of 2024, WisdomAI has expanded its client base dramatically, growing from just two enterprise customers to approximately forty. Notable clients now include Descope, ConocoPhillips, Cisco, and Patreon. According to CEO Mazumdar, the company is not only adding new customers but also seeing a rapid increase in usage among existing clients. Some organizations have doubled their engagement within two months, while one customer expanded from ten user licenses to 450, effectively covering nearly their entire workforce.
In the past six months, WisdomAI introduced an agentic functionality that provides real-time alerts to users regarding significant changes in the metrics they are tracking. Mazumdar illustrated this by describing an agent he personally configured to monitor product usage statistics and support ticket data. He emphasized that setting up this agent took roughly five minutes. Rather than delivering periodic reports, the system notifies him precisely when noteworthy developments occur.
Mazumdar expressed that this transforms traditional analytics from a static reporting tool into a dynamic and proactive resource. He believes the true value lies in shifting analytics from passive observation to active, timely insight, enabling businesses to respond instantly to critical changes.
(Source: TechCrunch)
