OpenLink Software

1/5 Data silos slow down things in the age of AI, as they did in prior eras. We help you break free. 🚀 From ODBC & JDBC connectors → Virtuoso Data Spaces → the OPAL AI Layer, we’ve built a vertically integrated portfolio to power AI for enterprises and individuals alike.

And here’s a presentation—generated by one of my agents—covering this shift in software development approaches. openlinksw.com/data//html/mvc…

And I forgot to attach an end-to-end screencast demo showing the creation and use of an AI agent using Markdown. It’s important to note that I use the OpenLink AI Layer (OPAL) as my AI agent creation environment, with Virtuoso as the deployment platform. This setup opens up…

Software Componentization in the Age of AI: How LLMs Are Reshaping Development linkedin.com/pulse/software… As usual, everything discussed is verified using live demos. The way software is developed and used is changing fast—and for the better! It’s AI agents all the way down. Once…

Session Links: 1. Basic session where the question’s text has an exact literal match in my Knowledge Graph: claude.ai/share/452a55ac… 2. More challenging test where the question’s text is only semantically similar to a question in my Knowledge Graph: claude.ai/share/d6889f55……

One more thing. Here are the links to questions in the knowledge showcased in this thread. 1. What is 'context' in the context of AI agents? -- linkeddata.uriburner.com/describe/?url=… 2. What is the guiding principle of good context engineering? -- linkeddata.uriburner.com/describe/?url=… 3. What is…

Agents.md is available from: github.com/OpenLinkSoftwa… And when the @OpenAI ecosystem opens up—just like we did with our DataTwingler CustomGPT—agents built and deployed in this manner will also be available for use. Remember: “Given a problem, there’s an App for…

As @OpenAI’s recent agentic e-commerce initiatives demonstrate: It has taken some time, but we’re finally transitioning to an Agentic Web that highlights the fundamental symbiosis between recent AI innovations—particularly natural language processing via LLMs—and structured data…

Traceability and observability are also features that are going to be a lot easier to utilize in the age of AI Agents. Here's a session log from the earlier screencast showcasing interactions with one of my AI Agents initialized via its loose coupling with my profile page and…

The workflow in play is straightforward: 1. Visit a page that catches my interest. 2. Use my AI Agent to generate notes, which are automatically uploaded to a Knowledge Graph. 3. I—or visitors to my profile—can ask questions via the AI Agent (functioning as my Personal…

Link to an example of a deceptively simple AI Agent that can perform SQL, SPARQL, SPASQL, or GraphQL queries against both local or remote data sources. 100% based on loose coupling of LLMs and Data Spaces (databases, knowledge bases/graphs, filesystems, APIs).…

Developers are crucial because they create the tools ultimately used by end-users, building on innovations in software development. That said, the new AI era fundamentally changes the nature of software development and use, as natural language becomes an integral part of the…

"..Here is my little interaction with an IRC Bot called "Phenny" (possibly mother of the Agent Web or Web of Agents or Web 4.0)" -- 2007 excerpt from my legacy blog. openlinksw.com/dataspace/doc/… I was wondering through old Semantic Web related posts when I stumbled upon that 😀

Cooking up a classic @OpenLink software innovation (OPAL, Virtuoso, and Data Access Drivers for ODBC and JDBC) storm—now that we’re finally in a new AI + Web era where AI Agents can be created and deployed using natural language (no need for WIMPY GUIs!). You simply describe your…

Now here’s a practical demonstration of the fundamental point I’m making about @danshipper’s insightful piece — in the age of AI Agents. I don’t just read an article, like it, or comment on it and move on to the next (as most people unfortunately do). I can ask my own AI Agent…

OPAL (@OpenLink AI Layer) enables the creation and deployment of powerful AI Agents that are loosely coupled with Data Spaces—databases, knowledge bases/graphs, filesystems, and APIs. Example: A SPASQL Agent (SPARQL inside SQL) that can be used via any MCP-compliant client,…

See OPAL and Sonnet 4.5 loose coupling effects in action via a Basic AI Agent configuration: •Static page: linkeddata.uriburner.com/chat/?chat_id=… •Animated playback: linkeddata.uriburner.com/chat/?chat_id=… Remember: An AI Agent created and deployed using OPAL is usable by any client application, service,…

So, @AnthropicAI released Claude Sonnet 4.5 earlier this week. As usual, I ran my own benchmark against my typical workflow: quickly digesting useful articles and generating notes for my Knowledge Graph. At the same time, I’ve been experimenting with new AI Agent configurations…

Here’s an example of what @Cloudflare is talking about, which I certainly wouldn’t characterize as an incorrect use of MCP. Essentially, you import an MCP that supports Server-Sent Events (SSE) or Streamable HTTP into a proxy service capable of publishing OpenAPI specs (YAML or…

💥Boom!💥 The developer story is being turbocharged with the @OpenLink AI Layer (#OPAL) and the Virtuoso Data Spaces platform, which enables semantic harmonization—through Knowledge Graphs—of information across databases, knowledge bases, filesystems, and APIs. This is how you…

Why should you learn and know about #ODBC (Open Database Connectivity)? It’s still the best API for developing and deploying #RDBMS-independent applications and services—without the compounding technical debt that comes from RDBMS-specific alternatives. @OpenLink has been…