Revo.pm

Salesforce Ventures just shared a great article featuring insights from Revo.pm and other AI leaders on selling to enterprise customers, a great read for people selling and deploying AI solutions to the enterprise. In a nutshell: - Build security & privacy into your foundation - Deep integrations and solid data sources are key to driving value - Focus on deep customization - it's your competitive edge - Make ROI clear and measurable Curious how Revo's enterprise agent can 10x your team's performance? DM me!

I love being mind-blown. And what a year it’s been—filled with surprises and technological advances that spark dreams of endless possibilities. Right now, I feel like I’m standing at the intersection of the past, the present, and the future. Philosophy aside, I’m thrilled to share three incredible solutions I recently discovered at the?Future of Work?networking event, organized by the Applied AI Association in San Francisco. Each one left me inspired about how AI is shaping the way we work. Today, I want to spotlight?REVO PM?(https://www.revo.pm). This powerful solution is a game-changer for product managers, acting as a one-stop shop to reduce the?busy work?that often consumes our time. Instead, it enables you to focus on where you bring the most value. Their motto, “Do more without doing more,” truly resonates. What stood out to me was the chance to chat with the founder, Mehdi Djabri, and hear his perspective on the future of work in the age of AI. It was refreshing to learn how much they prioritize their clients, offering thoughtful onboarding to ensure success. Stay tuned for more discoveries I’ll be sharing soon. Thanks for reading! #ProductManagement #AppliedAI #FutureOfWork

I will be speaking tomorrow at the Applied AI Association for their first event "The Future of Work with AI: Innovation & Transformation" at Intercom in San Francisco. Excited to share some our learnings at Iteration X building an ambient agent for product teams with Revo.pm. Come hang out with us and chat. Link below ??

How do you secure an AI agent with access to your entire company's data with many connected tools and users with varying rights and permissions? ?? I just shared some of our learnings and experiences securing Revo.pm at Okta in San Francisco for the amazing AI Tinkerers crowd. It's a fascinating security challenge that kept us up at night. Traditional ACLs and RBAC felt clunky for AI - we needed something more elegant. The breakthrough? Let Revo, the AI agent become its users. Through actor-based permission inheritance, our agent automatically operates with exactly the same access rights as whoever it's talking to. When in a group chat? It naturally works with only the data everyone has permission to see. No permission checks. No complex ACLs. No security theater. Some technical insights that made this work: - Security contexts flow like water through the system - Multi-user permissions resolve to set intersections - Vector stores inherit security boundaries - LLM responses can only reference allowed data - GraphQL queries execute in user context Real-world results after 4 months in production: ? Zero data leaks ? Millisecond overhead ? Seamless group interactions ? Natural permission boundaries ? Enterprise-grade audit trails Thank you for the great questions and engaging conversations, amazing AI engineering crowd and community??

Energizing discussion at the AI Tinkerers Palo Alto Meetup last night, where I shared our journey building Revo.pm, our AI Product Manager and Copilot for Product teams. We tackled a fascinating challenge: how to maintain coherent AI reasoning when handling multiple concurrent events. The core problem? LLMs need uninterrupted context to maintain their chain of thought - imagine trying to hold multiple conversations while someone keeps erasing your memory. Our solution: an event-driven architecture that treats AI reasoning as a first-class citizen, preserving context across interactions. Huge thanks to Joe Heitzeberg and Vasanth Mohan at SambaNova Systems for hosting, and to everyone who joined the technical deep-dive. The questions about race conditions and context management showed these challenges resonate across teams building AI agents. Want to learn more? Feel free to reach out!