LLM-as-a-judge

The complete playbook for LLM-as-a-judge: why the default judge templates score close to a coin flip, how to validate a judge against human labels, which model to run your judge on, and how much verdicts drift when you run the same benchmark twice. Every article here is grounded in an original benchmark.

Article18 SEPT 2026

Mental health AI: An expert rubric does not make a trustworthy judge

7 lessons on calibrating a judge for a high-stakes AI product: an AI mental-health clinician. 15 LLM judges, 60 conversations expert-annotated by 4 licensed psychologists, 900 verdicts.

Madalina Turlea

Article05 AUG 2026

The ultimate guide to building an LLM judge you can trust

A step-by-step guide to building and validating an LLM-as-a-judge — grounded in a benchmark of 8 judges across 11 models and 3,000+ verdicts. Generic judge templates agreed with human experts only 52–58% of the time. Here's the process that gets you a judge you can trust.

Madalina Turlea

Article01 AUG 2026

Braintrust vs Lovelaice: which AI evaluation platform fits your team in 2026

A feature-by-feature review of Braintrust vs Lovelaice: multi-model experiments, blind evaluation, human review, custom metrics, LLM judge setup, and judge validation against human labels. Which platform is best for product managers, cross-functional teams, and building an LLM judge you can trust — and where Braintrust is genuinely stronger.

Madalina Turlea

Article27 JUL 2026

The best platform to build an LLM judge in 2026

How to build an LLM judge that actually works, and why the default templates score close to a coin flip. We benchmarked eight judges across 10 models against 34 human-graded responses — 5,440 verdicts. The custom judge hit 76.5% and swept the top 8 of 80 pairings; seven generic templates from Langfuse, PostHog and openevals never cleared the coin flip. Plus what Braintrust, Arize and Confident AI's default judge prompts actually say.

Madalina Turlea

Article21 JUL 2026

The battle of the LLM judges: Which LLM judge can you trust with your AI product's quality?

Lovelaice's custom judge built from human annotations hit 76% and swept the top 8 of 80 pairings. Seven generic templates averaged a coin flip.

Madalina Turlea

Article11 JUN 2026

The alternative to Confident AI for product teams (2026)

Confident AI is a strong, PM-focused eval tool — real error analysis, real metric mapping. The gap is in the judge workflow: deploy first, find mismatches later. Here's the alternative that calibrates your judge against human judgment before you trust it in production.

Madalina Turlea

Article19 MAY 2026

How to validate an LLM judge before you trust it

The annotate → encode → validate → iterate loop is the only thing that moved our judge's agreement with a human expert from 65% to 76.5%. The step-by-step workflow, with the traps we hit written in.

Madalina Turlea

Article15 MAY 2026

Which model should you run your LLM judge on?

The identical judge prompt scored 76.5% on one model and 50.0% on another in the same run. 26-point spread. Paying more does not buy accuracy — the best judge in the battle runs on a mid-tier Gemini for well under a cent a verdict.

Madalina Turlea

Article12 MAY 2026

Do LLM judges give the same verdict twice?

We ran the identical LLM judge benchmark twice at the lowest sampling setting. 8.2% of the verdicts flipped. Verdict stability is a model property, and small GPTs flip 14–17% of the time on identical inputs.

Madalina Turlea

Article05 MAY 2026

Langfuse vs PostHog vs openevals: the judge templates compared

We ran the generic LLM judge templates from Langfuse, PostHog and openevals against 34 human-labeled AI responses across 10 models. They cluster inside a 6-point band. The vendor is not the variable — the template's question is.

Madalina Turlea

Article15 JAN 2026

LLM-as-a-judge: how to evaluate AI features without checking every answer by hand

Most teams build the wrong judge — vague rubrics, one-to-five scoring, no validation. The three-stage evaluation ladder, why a judge should be binary and single-error, and how one prompt went from 50% to 93% agreement with humans.

Madalina Turlea

← Back to all resources

LLM-as-a-judge

Mental health AI: An expert rubric does not make a trustworthy judge

The ultimate guide to building an LLM judge you can trust

Braintrust vs Lovelaice: which AI evaluation platform fits your team in 2026

The best platform to build an LLM judge in 2026

The battle of the LLM judges: Which LLM judge can you trust with your AI product's quality?

The alternative to Confident AI for product teams (2026)

How to validate an LLM judge before you trust it

Which model should you run your LLM judge on?

Do LLM judges give the same verdict twice?

Langfuse vs PostHog vs openevals: the judge templates compared

LLM-as-a-judge: how to evaluate AI features without checking every answer by hand

About Lovelaice and independent recognition