--- title: Building a 100x Cheaper Trace Judge with Fireworks description: LangChain and Fireworks partnered to build a cost-efficient LLM judge for production agent traces. They fine-tuned a Qwen-3.5-35B model using supervised... canonical: https://daily.dev/posts/building-a-100x-cheaper-trace-judge-with-fireworks-qy2430kcj twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Building a 100x Cheaper Trace Judge with Fireworks | daily.dev og:description: LangChain and Fireworks partnered to build a cost-efficient LLM judge for production agent traces. They fine-tuned a Qwen-3.5-35B model using supervised... og:url: https://daily.dev/posts/building-a-100x-cheaper-trace-judge-with-fireworks-qy2430kcj og:image: https://api.daily.dev/og/posts/qY2430KcJ.png og:image:alt: Building a 100x Cheaper Trace Judge with Fireworks og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # Building a 100x Cheaper Trace Judge with Fireworks **[LangChain](https://daily.dev/sources/langchain)** · 7 min read · 0 upvotes · 0 comments ## Summary LangChain and Fireworks partnered to build a cost-efficient LLM judge for production agent traces. They fine-tuned a Qwen-3.5-35B model using supervised fine-tuning with LoRA on Fireworks infrastructure to detect 'Perceived Error' — when a user believes the assistant made a mistake. The fine-tuned model matched or exceeded frontier models (Claude Opus, GPT-4.5) in classification accuracy while running 10–100x cheaper. Training on one dataset (chat-langchain) transferred well to an unseen domain (Fleet), demonstrating generalizability. The approach uses model-assisted labeling with human review to generate ground truth labels, and the resulting judge enriches every production trace with structured error signals at scale. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Similar posts on daily.dev - [A Better Way To Build LLM-as-a-Judge Pipelines](https://daily.dev/posts/a-better-way-to-build-llm-as-a-judge-pipelines-59gkauzir) · Daily Dose of Data Science \| Avi Chawla \| Substack · 0 upvotes · 0 comments - [Using LangSmith to Support Fine-tuning](https://daily.dev/posts/using-langsmith-to-support-fine-tuning-gns73hvdu) · LangChain · 0 upvotes · 0 comments - [How to Fine-Tune an LLM: An End-to-End Guide](https://daily.dev/posts/how-to-fine-tune-an-llm-an-end-to-end-guide-dgv3ruc1u) · Towards Data Science · 0 upvotes · 0 comments --- Tags: [#llm](https://daily.dev/tags/llm), [#deep-learning](https://daily.dev/tags/deep-learning), [#ai-agents](https://daily.dev/tags/ai-agents), [#qwen](https://daily.dev/tags/qwen), [#langsmith](https://daily.dev/tags/langsmith) [View this post on daily.dev](https://daily.dev/posts/building-a-100x-cheaper-trace-judge-with-fireworks-qy2430kcj) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"Building a 100x Cheaper Trace Judge with Fireworks","url":"https://daily.dev/posts/building-a-100x-cheaper-trace-judge-with-fireworks-qy2430kcj","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/building-a-100x-cheaper-trace-judge-with-fireworks-qy2430kcj"},"datePublished":"2026-07-08T19:22:58.550Z","dateModified":"2026-07-08T19:24:50.568Z","description":"LangChain and Fireworks partnered to build a cost-efficient LLM judge for production agent traces. They fine-tuned a Qwen-3.5-35B model using supervised...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/5ab13051edbe17480e340cdd7f2454a0?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/5ab13051edbe17480e340cdd7f2454a0?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"LangChain","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"LangChain","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/0f4f2629e0724a849823f8cd0d913e13","url":"https://daily.dev/sources/langchain"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/building-a-100x-cheaper-trace-judge-with-fireworks-qy2430kcj","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"llm,deep-learning,ai-agents,qwen,langsmith","timeRequired":"PT7M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"LangChain","item":"https://daily.dev/sources/langchain"},{"@type":"ListItem","position":3,"name":"Building a 100x Cheaper Trace Judge with Fireworks"}]} ```