Connecty AI

'Explainability' is the biggest risk to the success of AI data agents—not accuracy. Most AI agents today rely on custom RAG (Retrieval Augmented Generation), which inherently suffer from the black-box nature of LLMs. The issue isn’t just about accuracy - it's that we don't know WHY a specific outcome was generated. When it comes to complex data questions, this becomes a real problem. Take for example a sales person may ask: “I need a breakdown of our suppliers. Specifically, I want to know how much revenue each supplier is generating, which customers contribute the most to their revenue, and what product categories they sell the most. Also, can we calculate what percentage of a supplier’s revenue comes from its top customers? Finally, I’d like to see each supplier’s market share compared to the total revenue across all suppliers. Let’s sort by the biggest suppliers first.” Here, traditional RAG agents repeatedly miss the dependency of a sub-question on parent, or make wrong assumptions without clarifying, resulting in an opaque incorrect output. Most data teams that we work with don’t just want the answer and blindly assume it’s correct. They want to understand the journey and thought process from question to SQL to insight. Here are some tips for anyone building AI agents for structured data: ? Break down intent into sub-intents, handling 1:many dependencies ? Solve each sub-intent in the context of its parent ? Explain each step involved in custom RAG, e.g. context filtering ? Trace and explain the full reasoning path ? Let users debug, fix, and trust At Connecty, we’re pioneering an explainability layer that breaks down every step of intent, deep reasoning, and context selection in a human-comprehensible format—while giving users the ability to refine and provide feedback. That’s how we’re winning the trust of our users. We’re building the world’s most explainable Agentic AI platform for data, making answers explainable—not just accurate. Stay tuned! #AI #DataAnalytics #LLM #Explainability #SQL #Connecty #EnterpriseAI #HumanInTheLoop

In May, Insight Partners discussed an impending shift in enterprise architectures toward AI Agents. In the six months since that article came out, we have monitored actual agentic deployments across companies, noticing the differences in use case complexity, the tradeoff of buying function-specific Agents versus building custom agentic workflows, and the variety in how value is measured and attributed. We spoke to builders, enterprises, end users, and researchers to get a deeper understanding of where Agents stand today, including: ?? The state of the AI Agent ecosystem ?? An updated AI agents market map ?? Agent end-user deployments: Examples, challenges, and considerations ?? The economics of agents ?? Considerations for enterprises ?? Considerations for builders Written by Managing Director Praveen Akkiraju, VP Sophie Beshar, Director Ash Tutika, Hunter Korn, Ethan DeSilva, and Christopher Gyra, see how the AI Agents space has already evolved: https://lnkd.in/e48KNgeM

Absolutely energized after attending the Data Science Salon at Google San Francisco office, hearing to brilliant ML/data science minds from Linkedin, Google, IBM, Pinterest, SAP, Headspace and Disney Vyoma Gajjar, Jodi Blomberg, Raghavan Muthuregunathan, Swagata Ashwani, Shivika K Bisen Ari Heljakka, Akhil Chaturvedi, Sayan Maity, PhD and more. Really appreciated exchanging thoughts with some of you on autonomous agents' limitations and how Connecty AI is pioneering agentic workflows for structured data, augmenting humans expertise and not replacing it! A huge shoutout to Anna Anisin for putting together such an impactful event! ?? Looking forward to attending 'AI Agents + Knowledge Graphs => Intelligent Systems' this Thursday, November 14th in San Francisco. If you’re attending, let’s connect! #DataScience #GenAI #DataAgents #Innovation

There’s a lot happening for Connecty AI lately—and we’re just getting started! ?? Today, we’re at Web Summit meeting amazing founders, innovators, and curious minds. The conversations have been incredible and packed with fresh ideas. If you’re here too, come find us at stand A249 in the AI & Machine Learning section. We’d love to show you what Connecty AI is all about! But please come say hi soon—our voices are hanging on by a thread! ?? Photo taken by Michael Bernardi, our neighbour from ORPIVA AI!