Notes on building, evaluating and operating AI agents.https://www.thaliq.com/enhttps://www.thaliq.com/blog/a-worse-version-never-reaches-production/https://www.thaliq.com/blog/a-worse-version-never-reaches-production/This is how Thaliq’s eval gate works: every change to an agent is measured before it is published. If the score doesn’t reach the threshold, the version already serving your customers stays where it is.Wed, 14 Oct 2026 00:00:00 GMTguidehttps://www.thaliq.com/blog/how-your-agent-knows/https://www.thaliq.com/blog/how-your-agent-knows/A model doesn’t know your return policy. Retrieval finds the right passages of your own content and hands them to the model with the question. Here is how Thaliq does it, step by step.Sun, 28 Jun 2026 00:00:00 GMTengineeringhttps://www.thaliq.com/blog/anatomy-of-an-agentic-loop/https://www.thaliq.com/blog/anatomy-of-an-agentic-loop/An agent isn’t a prompt: it’s a loop that reasons, calls tools, observes and repeats until it can answer. Here is that loop, running.Wed, 24 Jun 2026 00:00:00 GMTengineeringhttps://www.thaliq.com/blog/tokens-at-cost/https://www.thaliq.com/blog/tokens-at-cost/Some platforms resell you the model with a margin on every token. A Thaliq paid plan is a monthly fee plus the tokens you use, at what they cost us, with no margin for now.Thu, 18 Jun 2026 00:00:00 GMTproducthttps://www.thaliq.com/blog/where-your-data-runs/https://www.thaliq.com/blog/where-your-data-runs/The same platform can run in two places: Thaliq’s cloud, in AWS, or your own infrastructure. Here is what lives where, and what leaves.Wed, 10 Jun 2026 00:00:00 GMTsecurity