Systems | Development | Analytics | API | Testing

Training models for physical robots is harder than training a typical model. The data has to be collected by hand through teleoperation, every change has to be tested on real hardware, and the loop from data to deployment runs constantly. In a recent proof of concept with a Singapore government agency, ClearML, Dell Technologies, and Hugging Face’s LeRobot framework turned that high-touch, manual process into an automated pipeline.

Twenty years of boots-on-the-ground testing experience reveals a clear pattern: the industry has moved from tracking manual test cases in Excel sheets, to managing Selenium Grid configurations, to watching algorithms generate scripts in seconds. Right now, if you are in a managerial role, your feeds are absolutely flooded with pitches promising that.

You ask your agent a question. The answer is slightly off. You point out the gap. Spotter fixes it, and that fix doesn't disappear when the session ends. Your team doesn't re-explain the same thing tomorrow. The next analyst doesn't start from scratch. The correction stays, and the work gets better from here. That's what memory makes possible. Not just for you. For everyone who comes after.

*Kong makes every MCP client and server work with Enterprise-Managed Authorization, whether they speak the protocol or not.* The MCP demo impressed the room. Then someone asked how 5,000 employees would connect to 40 MCP servers, and the answer was: one OAuth consent screen at a time. Per user. Per server. No central policy, no unified audit trail, and nothing stopping a personal account from getting wired into a work tool.

AI Transport v0.4.0 includes changes to optionally support database hydration. Some applications may wish to store AI conversation history in an external store, such as a database. AI Transport's support for database hydration allows applications to reconcile that stored history with the live activity in the AI session. When using database hydration, your application persists messages for completed runs to the database.

LLM provider switching went from a theoretical concern to an operational emergency in June 2026, when Anthropic disabled Claude Fable 5 and Mythos 5 following a US government directive . The shutdown was swift, with access suspended just days after the models launched. Enterprises that had built production workflows around those models lost access overnight. The event was a wake-up call, but the underlying risk had been building for years.

In late June 2026, two of the most widely used AI agent platforms were compromised within the same week. Langflow disclosed a critical unauthenticated remote code execution flaw. Dify, powering over one million applications, revealed four vulnerabilities that exposed private conversations and internal APIs across tenant boundaries. These weren't theoretical risks. They were production exploits hitting real infrastructure.

The agentic AI conversation is too obsessed with the front door. Customer-facing agents get most of the attention because they are visible.