The great AI unbundling is marking a shift from one-size-fits-all solutions to a LEGO-like world where developers can pick the best AI models and tools, mixing and matching components to build systems that perfectly fit their needs.
In this article, I'll take you through the step-by-step process of building an MCP server and connecting it to a custom solution on LangGraph (client).
What if AI could free us from endless paperwork? Through Tukki's legal-tech journey, learn how proper digital foundations can transform administrative tedium into meaningful work.
Bringing AI into a company is like hiring a new employee — one that’s demanding, detail-oriented, and can’t rely on intuition. AI, like any new team member, needs everything explained clearly and thoroughly.
This process often reveals that our knowledge isn’t as well-structured or organized as we thought. Bringing AI into our operations makes us take a closer look at our processes and documentation, showing us gaps and inefficiencies we might be missing.
Want your AI to deliver accurate answers every time? There's a critical element many teams overlook.
A Ground Truth Dataset is your foundation for reliable LLM evaluation. While some metrics don't need it, having this human-validated reference data is essential for truly understanding if your AI is performing as intended. Read on to discover why your Subject Matter Experts need to care about Ground Truth, and how to build it right.
After OpenAI announced the public release of Swarm, a library designed to build reliable multi-agent systems, I decided to dive in and explore its potential by integrating it into a real-world application.