Read Machbase technical articles.

Machbase NEO V8.5 ushers in the era of AI agents Is your data piling up while you struggle to use it to automate your work? Until now, database management systems (DBMSs) have followed a human-centered model: people had to write SQL and explore data manually before anything could happen. That era is changing. As AI agents become the users of databases, Machbase is at the center of this transformation. At 10 a.m. on April 27, 2026, Machbase NEO V8.5 officially launches: next-g

Anyone responsible for data at an industrial site has probably faced questions like these: “How many gateways and middleware packages do I need to install just to collect PLC data?” “If a transfer between servers stops midway, how can we recover the missing data?” “Collection servers, database servers, analytics engines… managing all of them is becoming overwhelming.” Industrial sites are getting smarter, but the way we handle their data remains complicated. And complexity le

We’re excited to share the innovative update so many of you have been waiting for! Machbase NEO V8.5 goes beyond faster data storage: it gives the database engine “eyes” through Black-box and a “brain” through LLMs. Bring previously disconnected video and sensor data together, and discover a new way of working—gaining insights through conversation with AI, without complex queries. NEO V8.5 App Packages, Part 3 Black-box & LLM *Click the image to watch the corresponding video.

We have entered an era in which you can ask about data in natural language and let AI draw charts and even build dashboards for you. Machbase Neo’s neo-pkg-llm-chat package is an LLM agent that makes this experience possible. If you have used an earlier version, however, you may have encountered frustrations like these: “I connected a local LLM through Ollama, but it keeps calling the wrong tools…” “Why are the API costs higher than I expected?” “When several people use it to

The last weakness of 2.x was code written directly by AI. Version 2.0.0 greatly reduced failures by separating features according to intent and automatically correcting invalid input. But one class of problem remained: chart code written directly by AI. Creating a chart requires query code behind the scenes, and that code contains many pitfalls. The AI repeatedly made the same mistakes in handling time values and applying aggregation and interval rules. Lightweight local mode