Journeys by RenderDraw: Transform slides into immersive 3D experiences https://t.co/E1tYGYTUJD via @producthunt
We built Journeys because we were tired of presentations that lose the room.
You know the moment: eyes glaze over, phones come out, and the story dies.
So we asked a simple question:
What if a presentation could feel alive?
Journeys turns static decks into immersive, interactive experiences with AI built into the workflow.
Not to replace your voice, but to amplify it.
What we’re especially proud of right now:
Import from PowerPoint, PDF, or URL and transform fast
AI assistant + story tools + JOE scoring to improve clarity and impact
Real presentation tools: camera paths, flyovers, spotlight/presenter controls
Session analytics and heatmaps so you know what actually landed
Real-time collaboration, sharing, embeds, and developer integrations (API/webhooks/MCP)
Special thanks to @mrdoob and @threejs, @claudeai and @GeminiApp
Salesforce’s native debug logs come with limits such as size caps, truncation, and short retention. Nebula Logger provides a more powerful and scalable way to track errors across Flow, Apex, and LWC.
Atlas Can shares a practical guide to Nebula Logger.
https://t.co/Cvzt9wFZAd
Salesforce's native debug logs work - until they don't. Truncation limits, a cap of users at a time, no structured querying, and System.debug() that only works in Apex. Once you're debugging complex Flows or LWC, you start hitting walls fast.
But this is exactly where the Salesforce ecosystem shines. I wrote a practical guide about my favorite logging solution, Nebula Logger on @SalesforceBen.
https://t.co/XcQst94nIM
Enjoyed the article? If you want to support this series and high signal content without fluff, sharing it with your network is the best way.
Turning Salesforce data into reliable AI answers isn’t easy—the gap often lies in retrieval and prompting.
Atlas Can explains how to build a custom retriever in Einstein Studio, test it in Prompt Builder, and connect it to an Agentforce Service Agent.
https://t.co/EJZ6oFDuC8
In this part of the series, we are exploring how to create a custom retriever that uses the web crawler connection from part 1, connect it to the service agent and verify our results using the prompt template for our RAG based service agent.
Enjoyed the article? If you want to support this series and high signal content without fluff, sharing it with your network is the best way.
https://t.co/mjEWZNKFMZ
Agentforce just hit 11.14 TRILLION tokens served - that's a real force of real agents! Real use & real adoption. Powering agents for sales, service, marketing, commerce, Tableau, & Slack across the enterprise. Mahalo to our Ohana for their epic velocity after only a year in service, and you're making the Agentic Era real! ❤️🤖
In a time where customer service expectations are higher than ever, providing accurate, real-time information is crucial.
Atlas Can explores how to build RAG-powered agents with Salesforce Data Cloud and Agentforce.
https://t.co/170fTWrzjV
#Salesforce
Friends and colleagues,
Agentforce Atlas Reasoning Engine, ChatGPT Atlas, 3I/ATLAS. I keep seeing my name everywhere, which can obviously mean one thing: I must write about agentic workflows in 2026.
Joking aside, here's the first part of a step-by-step guide on creating a RAG-powered agent that connects to and scrapes an external website using a Data Cloud web crawler and a custom retriever in Salesforce on @SalesforceBen.
Based on real-world scenarios and my own experience, I will be sharing my learning journey in this exciting AI era completely free, as I wish everyone a happy and healthy new year.
Enjoyed the article? If you want to support this series and high signal content without fluff, sharing it with your network is the best way.
https://t.co/KJl3zAoRDR
@karpathy Perhaps it's a case of convergent evolution, different models arriving at similar solutions for effective communication? Maybe there are universal language rules underpinning this.
@bate5a55@arankomatsuzaki Are there specific types of models or datasets where this predictive approach for optimal model size has been particularly effective, or does its applicability vary widely across different machine learning scenarios?
Salesforce's newest version of Agentforce, code named Atlas, resolves 90%+ of all customer inquiries for our top healthcare, financial services, payments, travel & entertainment clients—over double the success rate of competing Agents. State-of-the-art results. See it at Dreamforce. ❤️