# ai quality

Published articles for ai quality.

This is one page of public article previews, not the complete archive. Follow Next page to continue. Summaries are not the original full articles.

## LLM-судья для нейроразбора резюме на hh

DevFeed: [LLM-судья для нейроразбора резюме на hh](<https://devfeed.tech/articles/llm-hh-30669.md>)

Original publisher: [Read original article](<https://habr.com/ru/companies/hh/articles/1050174/>)

Author: Mentalitet (hh.ru)

Published: 2026-06-22T05:31:17Z

Content type: article

Language: ru

Sources: [HeadHunter RU](<https://devfeed.tech/sources/headhunter-ru.md>)

Topics: [Large Language Model](<https://devfeed.tech/topics/llm.md>), [Machine learning](<https://devfeed.tech/topics/machine-learning.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ai-quality](<https://devfeed.tech/tags/ai-quality.md>), [eval](<https://devfeed.tech/tags/eval.md>), [hh-ru](<https://devfeed.tech/tags/hh-ru.md>), [hr](<https://devfeed.tech/tags/hr.md>), [llm](<https://devfeed.tech/tags/llm.md>), [llm-10b5a83db3c5](<https://devfeed.tech/tags/llm-10b5a83db3c5.md>), [llm-as-a-judge](<https://devfeed.tech/tags/llm-as-a-judge.md>), [llm-as-judge](<https://devfeed.tech/tags/llm-as-judge.md>), [quality-assurance](<https://devfeed.tech/tags/quality-assurance.md>), [tag-61cd5a476b1d](<https://devfeed.tech/tags/tag-61cd5a476b1d.md>), [tag-a939ae104bb0](<https://devfeed.tech/tags/tag-a939ae104bb0.md>), [tag-b6914c0b0244](<https://devfeed.tech/tags/tag-b6914c0b0244.md>), [tag-d96e238f1419](<https://devfeed.tech/tags/tag-d96e238f1419.md>), [tag-e7e38deac50c](<https://devfeed.tech/tags/tag-e7e38deac50c.md>)

### AI overview

This article describes how hh.ru designed an LLM judge for its resume-analysis feature in an AI assistant. It explains why reliable evaluation requires its own rubrics, datasets, quality metrics, operating-cost considerations, and controls for hallucinations and interpretation errors. The system assesses candidates against criteria relevant to a specific position, which may be extracted from a vacancy or defined by an employer.

### Source excerpt

Создать LLM-судью легко. Гораздо сложнее сделать так, чтобы его оценкам можно было доверять. Мы убедились в этом на практике при разработке нейроразбора резюме для ИИ-помощника hh.ru. Быстро выяснилось, что хороший LLM-судья -- это отдельный продукт со своими рубриками, датасетами, метриками качества и стоимостью эксплуатации. Меня зовут Женя Орлов, я LLM Eval Lead. В этой статье расскажу, как мы проектировали систему оценки для нейроразбора резюме, почему отказались от наивных подходов и какие выводы сделали по ходу разработки. Читать далее

## Kelet: Building a durable workflow to diagnose AI agent quality failures

DevFeed: [Kelet: Building a durable workflow to diagnose AI agent quality failures](<https://devfeed.tech/articles/we-built-a-durable-agent-that-debugs-durable-agents-36096.md>)

Original publisher: [Read original article](<https://temporal.io/blog/we-built-a-durable-agent-debugs-durable-agents>)

Author: Almog Baku

Published: 2026-06-18T00:00:00Z

Content type: article

Language: en

Sources: [Temporal Blog](<https://devfeed.tech/sources/temporal-blog.md>)

Topics: [agent observability](<https://devfeed.tech/topics/agent-observability.md>), [Orchestration](<https://devfeed.tech/topics/orchestration.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [AI-assisted coding](<https://devfeed.tech/topics/ai-assisted-coding.md>)

Tags: [agents](<https://devfeed.tech/tags/agents.md>), [ai-agents](<https://devfeed.tech/tags/ai-agents.md>), [ai-quality](<https://devfeed.tech/tags/ai-quality.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [bugs](<https://devfeed.tech/tags/bugs.md>), [community](<https://devfeed.tech/tags/community.md>), [don-t](<https://devfeed.tech/tags/don-t.md>), [orchestration](<https://devfeed.tech/tags/orchestration.md>), [traces](<https://devfeed.tech/tags/traces.md>)

### AI overview

This article explains how Kelet diagnoses recurring quality failures in production AI agents by processing sessions, accumulating hypotheses, and reasoning across many occurrences. It describes why durable workflows are needed instead of a single agent loop or LLM call.

### Source excerpt

Almog Baku, co-founder and CEO of Kelet AI built an AI that diagnoses failing AI agents in production. Here's how he did it and why it needed durable workflows.