Anaconda, Inc.

Happy to share what I've been working on with the R&D team at Anaconda, Inc. : 🤖 Anaconda Agent Studio is out in Beta NOW 🤖 - Install Anaconda Desktop. - Turn on Agent Studio in settings (from profile menu) - Launch from the Agent Studio button on the left nav This is an experimental feature. We're shipping early to get community feedback so we can iterate and improve. - Build agents with an easy UI or just ask the assistant what you want your agent to do and have it build one for you! - Use your own AI provider (OpenAI, Anthropic, etc) or a local AI model running on Anaconda Desktop - Give your agents capabilities - plugins, MCP servers, skills and light-weight Python functions - this latter one is probably most powerful as you can be very specific about those tasks you need done but for which you don't have an MCP server. Do all of this with the UI, or just ask the assistant to do it for you. - Build Tool Servers - Python functions running on a local process that your agents can use. Again, use the UI, or ask the assistant to do it for you. Launch your agents on a local process or in a Docker container. Give it a go and tell me what you like or don't like. We want to build something the community wants to use! --- For a test yesterday I built an email summarizer using a local AI server that reads and summarizes emails across the 5 accounts I have set up in Outlook. Local AI = no subscription cost, no inference cost, data stays local. Interested in an informal meetup to chat about it? Let me know! #anaconda #agents #ai

Exciting day at Anaconda, Inc. — we're announcing AI Catalyst, our new enterprise AI development suite, available for enterprises running on Amazon Web Services (AWS). In the open source AI space, the challenge is clear: there are millions of models, but figuring out which ones are production-ready, secure, and optimized for your infrastructure? That's where teams lose momentum. AI Catalyst addresses this head-on. We've built a curated catalog of models that come with:  - Complete AI Bills of Materials and security assessments  - Pre-applied quantization for better performance at lower cost  - Standardized benchmarks so you can make data-driven decisions  - Built-in governance policies that adapt to your compliance requirements And the deployment story is flexible—prototype locally, deploy to GPU endpoints in your Amazon Virtual Private Cloud, all with the same consistent experience. I'm really excited about what this enables for enterprise teams. We're making it possible to move from experimentation to production without the usual security and infrastructure bottlenecks. Curious to hear thoughts from others working on enterprise AI initiatives. https://bit.ly/44g4Akv