Field notes on testing conversational AI agents — multi-turn failures, tool-calling, hallucinations, and how to catch them before your customers do.https://voxli.io/en-ushttps://voxli.io/blog/a-better-model-can-make-your-agent-worse/https://voxli.io/blog/a-better-model-can-make-your-agent-worse/A stronger model with higher scores looks like a free upgrade, until the agent that worked last week starts getting things wrong, quietly. Here is what happened when we ran one agent on two frontier models and changed nothing else.Wed, 17 Jun 2026 06:41:00 GMTAgent ReliabilityConversational AIAI Quality AssuranceAI Agent TestingModel BehaviorAI AgentsVoxlihttps://voxli.io/blog/upfront-information-dump/https://voxli.io/blog/upfront-information-dump/A customer opens your support agent with this:Tue, 26 May 2026 11:05:00 GMTFailure ModesAI AgentsConversational AIMahey Qadirhttps://voxli.io/blog/mid-conversation-tangent/https://voxli.io/blog/mid-conversation-tangent/A customer is halfway through a return flow with your agent. They've shared the order number, the item and reason for the return. They then pause to ask: "Wait, do you offer…Fri, 15 May 2026 14:15:00 GMTVoxlihttps://voxli.io/blog/multi-turn-failures/https://voxli.io/blog/multi-turn-failures/Most agent failures we see in pilots don't show up on prompt evals.Mon, 27 Apr 2026 14:46:00 GMTAgent ReliabilityAI AgentsAI Agent TestingAI Quality AssuranceConversational AISupport AgentVoxlihttps://voxli.io/blog/the-10-minute-test-that-stops-your-agent-from-canceling-real-orders/https://voxli.io/blog/the-10-minute-test-that-stops-your-agent-from-canceling-real-orders/Last week a failed tool call caused GPT-5.4-mini to cancel a real order simply because a customer asked a question involving cancellation. Here's a quick test that catches it.Tue, 21 Apr 2026 09:34:00 GMTVoxlihttps://voxli.io/blog/expertise-ai-teams-up-with-voxli/https://voxli.io/blog/expertise-ai-teams-up-with-voxli/Expertise.ai is a known disruptor in the AI space, building AI sales agents that guide prospects through personalized flows. Here's how Voxli untangled their testing workflow.Thu, 16 Apr 2026 12:06:00 GMTCase StudyCustomer StoryMahey Qadirhttps://voxli.io/blog/ai-agents-tool-handling/https://voxli.io/blog/ai-agents-tool-handling/Recently, to assess AI Agent performance with tool calls, we executed the same multi-turn conversation across the three tiers of OpenAI's GPT-5.4: standard, mini, and nano.Tue, 14 Apr 2026 08:48:00 GMTAI Agent TestingAI AgentsMahey Qadirhttps://voxli.io/blog/testing-for-speculation-using-voxli/https://voxli.io/blog/testing-for-speculation-using-voxli/In our last post we covered the risks of agent speculation. Today we look at how to set up Voxli to catch those speculations — using a feature called Hallucination detection.Thu, 02 Apr 2026 11:47:00 GMTAI Agent TestingHow-to-guideMahey Qadirhttps://voxli.io/blog/risks-of-agent-speculation/https://voxli.io/blog/risks-of-agent-speculation/It's no surprise that hallucinations are a common known failure during agentic AI testing. The agent starts to overpromise, begins to fabricate answers and even claims that it…Fri, 27 Mar 2026 15:26:00 GMTAI AgentsAI Agent TestingLLM TestingModel BehaviorReasoning ModelsAgent ReliabilityConversational AIAI Quality AssuranceSupport AgentVoxli