Technology Radar

We’ve seen significant movement in the observability space, driven by the growing complexity of distributed architectures. While observability has long been essential, it continues to evolve alongside the rest of the software development ecosystem. One emerging focus is LLM observability, a critical piece in operationalizing AI. We’ve seen a surge in tools for monitoring and evaluating LLM performance, including Weights & Biases Weave, Arize Phoenix, Helicone and HumanLoop. Another trend is the integration of AI-assisted observability, where tools leverage AI to enhance analysis and insights. Additionally, the increasing adoption of OpenTelemetry is fostering a more standardized observability landscape, enabling teams to remain vendor-agnostic and more flexible in their tooling choices. Many leading observability tools — such as Alloy, Tempo and Loki — now support OpenTelemetry. The rapid innovation in observability tools demonstrates growing industry awareness of observability’s importance, creating a cycle where evolving practices and technologies reinforce each other.

Big data has long been a key concern for the industry, but conversations around this edition of the Radar centered not so much on size but more specifically on handling rich and complex data. With the increasing presence — and importance — of unstructured data in the enterprise, ensuring data is effectively managed and packaged so it can be successfully leveraged for everything from AI applications to customer analytics is today vital for businesses.

This is reflected in many blips: from vector database tools to analytics products like Metabase, it’s astonishing how much of the software ecosystem is being driven by what we want and need to do with data. It’s not just about tooling though — in this edition we noted data product thinking, a framework that encourages teams to apply the principles of product thinking to the analytic parts of their ecosystem. The emergence of data product thinking is, to some extent, a result of the ongoing challenge of properly leveraging data (something that’s been talked about for years, well before the rise of AI) — the fact that it has found its way into the spotlight (and our Radar conversations) demonstrates that the need for data discipline is now as great as it has ever been. Without it, organizations may struggle to innovate and could well be at a commercial disadvantage in the medium and long term.