Notifications
Clear all
Topic starter
20/08/2026 6:10 pm
TL;DR: AI quality improves when teams turn production traces into datasets, then use code-based scorers for deterministic checks and LLM-as-a-judge scorers for nuanced outputs, according to Braintrust. The core shift is from shipping AI on vibes to treating evaluation as infrastructure, because agent failures often only appear when tool use, retrieval, or synthesis breaks in production.
NHIMG editorial — based on content published by Braintrust: How to evaluate LLMs and AI agents in production
Questions worth separating out
Q: What breaks when AI prompts are changed without evaluation?
A: The system may appear to work in a demo while silently degrading in production.
Q: Why do AI agents need step-level evaluation as well as end-to-end testing?
A: End-to-end success can hide fragile behaviour in the middle of the workflow.
Q: How do security teams know if automated AI evaluation is actually working?
A: Look for stable agreement with human reviewers, low sensitivity to answer order, and consistent scores across repeated tests on the same inputs.
Practitioner guidance
- Instrument every production trace Capture prompts, retrieval spans, tool calls, and final outputs so each AI interaction can be replayed as a test case when quality drops or behaviour changes.
- Define separate scoring rules for deterministic and subjective quality Use code-based scorers for schema, format, and factual checks, then apply LLM-as-a-judge scoring only to qualities such as clarity, tone, or reasoning.
- Score agent steps, not just final answers Evaluate tool selection, argument construction, and handling of tool output independently so a successful final answer does not hide a risky intermediate decision.
What's in the full article
Braintrust's full article covers the operational detail this post intentionally leaves for the source:
- The exact data, task, and scorer pattern used to structure evals for production AI systems.
- Concrete examples of how trace logs are turned into datasets and reused in offline testing.
- The workflow for combining human review with automated scoring in AI product teams.
- The practical mechanics of agent evaluation across tool choice, argument quality, and output synthesis.
👉 Read Braintrust's guide to evaluating LLMs and AI agents in production →
AI agent evaluation in production: are your controls keeping up?
Explore further
20/08/2026 6:46 pm
Trace-based evaluation is now a governance control, not just an engineering practice. AI programmes that cannot replay production behaviour cannot reliably explain why an agent chose a tool, exposed an output, or degraded in quality. That leaves reviewers with symptoms instead of evidence. For identity and security teams, the practical takeaway is that traceability has become part of AI governance, because runtime decisions are where policy is either enforced or bypassed.
A question worth separating out:
Q: How should organisations govern external tools used by AI agents?
A: Organisations should review external tools as security inputs, not convenience features. Each tool needs ownership, approval, metadata inspection, and ongoing monitoring for hidden instructions or unexpected behaviour. If an AI agent can act on a tool, then the tool’s provenance and control status should be governed like any other sensitive integration.
👉 Read our full editorial: Evaluating LLMs and AI agents in production needs trace-based tests
