I just published Your MCP Server Has a Token Tax. mcp-xray tells you exactly how much. medium.com/p/your-mcp-server…
Who said vibe coding "on the weekend" (the city never sleeps) can't be fun? Now I have also added a full OrionBelt-Dremio-in-the-loop demo in the actual v2.11 release to show how the OrionBelt Semantic Sidecar can work
lnkd.in/dvTnSuMd
#semanticsidecar #orionbelt #dremio
Freshness-driven result cache is now shipped with OrionBelt Semantic Layer v2.2.0:
github.com/ralfbecher/orionb…
🤖 Made with AI
Every semantic-layer cache I've seen declares freshness in the wrong place. On the cube. On the saved query. On the datagroup. Not on the source table, which is the only place where refresh actually happens.
medium.com/p/freshness-inher…
🚀 MCP sampling now live in OrionBelt Chat v1.1.4 ↔ OrionBelt Analytics v1.5.0.
Official MCP clients matrix: 17 of 113 support sampling. Claude Desktop, ChatGPT, Cursor, Cline, Windsurf, none of them. But we do.
Jebi ga, finally. :D
github.com/ralfbecher/orionb…
#MCP #AgenticAI
Spent the week writing something nobody asked for: 5 honest comparison pages for OrionBelt Semantic Layer vs dbt, Malloy, LookML, Cube, AtScale.
Each includes a “what OBSL is missing” section — all-green vendor matrices are useless.
ralforion.com/orionbelt-sema…
MetaDada retweeted
🚀 DeepSeek-V4 Preview is officially live & open-sourced! Welcome to the era of cost-effective 1M context length.
🔹 DeepSeek-V4-Pro: 1.6T total / 49B active params. Performance rivaling the world's top closed-source models.
🔹 DeepSeek-V4-Flash: 284B total / 13B active params. Your fast, efficient, and economical choice.
Try it now at chat.deepseek.com via Expert Mode / Instant Mode. API is updated & available today!
📄 Tech Report: huggingface.co/deepseek-ai/D…
🤗 Open Weights: huggingface.co/collections/d…
1/n
MetaDada retweeted
Watchdogs say the previous CEO model suspiciously started slowing down immediately after the announcement
Unified Star Schema: I think this concept is outdated in the agentic AI world:
linkedin.com/feed/update/urn…
#agenticai #analytics #businessintelligence
🚀 Reaching the next level in AI-Powered Analytics.
Combining schema discovery, ontology-based knowledge modeling, semantic enrichment, and GraphRAG, so AI doesn't just query your data, it ✨ *understands your database*. medium.com/p/orionbelt-analy…
Working with NanoClaw in Docker for a while now. The setup was a bit of a hassle. However, it now runs perfectly on my M3 Max, and soon I will migrate to a Pi5 with almost no power consumption.
What is your experience with #NanoClaw?
github.com/qwibitai/nanoclaw
🤔 What if your AI never hallucinated a table name again?
Text-to-SemanticLayer > Text-to-SQL.
I built an open-source engine to prove it 👇
medium.com/@irregularbi/buil…
🚀 OrionBelt Semantic Layer now supports bidirectional OSI integration! 🔄
Import & export with validation, lossless ai_context, and a live UI. 5 SQL dialects · REST API · MCP server 🤖
💻 github.com/ralfbecher/orionb…
#SemanticLayer #OSI #SQL #MCP #AI #OpenSource #DataEngineering
MetaDada retweeted
Peter Steinberger is joining OpenAI to drive the next generation of personal agents. He is a genius with a lot of amazing ideas about the future of very smart agents interacting with each other to do very useful things for people. We expect this will quickly become core to our product offerings.
OpenClaw will live in a foundation as an open source project that OpenAI will continue to support. The future is going to be extremely multi-agent and it's important to us to support open source as part of that.
Did a small contribution to the Open Semantic Interchange (OSI) #semanticlayer repo:
github.com/open-semantic-int…












