# How we used DSPy to turn AI evaluations into better responses in Dash chat

DevFeed: [How we used DSPy to turn AI evaluations into better responses in Dash chat](<https://devfeed.tech/articles/how-we-used-dspy-to-turn-ai-evaluations-into-better-responses-in-dash-chat-177.md>)

Original publisher: [Read original article](<https://dropbox.tech/machine-learning/how-we-turned-ai-evaluations-into-better-responses-in-dash-chat>)

Author: Yasmin McDowell,Lawrence Good,Ilya Yakovlev,Andrew Cheung,Binoy Dash,Simran Jumani,Dmitriy Meyerzon

Published: 2026-06-25T16:30:00Z

Content type: article

Language: en

Sources: [Dropbox Tech Blog](<https://devfeed.tech/sources/dropbox-tech-blog.md>)

Topics: [DSPy](<https://devfeed.tech/topics/dspy.md>), [AI Chat](<https://devfeed.tech/topics/ai-chat.md>), [Large Language Model](<https://devfeed.tech/topics/llm.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [Optimization](<https://devfeed.tech/topics/optimization.md>), [Algorithms](<https://devfeed.tech/topics/algorithms.md>), [Open Source](<https://devfeed.tech/topics/open-source.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [dash](<https://devfeed.tech/tags/dash.md>), [dspy](<https://devfeed.tech/tags/dspy.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [llm](<https://devfeed.tech/tags/llm.md>), [open-source](<https://devfeed.tech/tags/open-source.md>), [optimization](<https://devfeed.tech/tags/optimization.md>), [performance](<https://devfeed.tech/tags/performance.md>), [reasoning](<https://devfeed.tech/tags/reasoning.md>), [responses](<https://devfeed.tech/tags/responses.md>), [tools](<https://devfeed.tech/tags/tools.md>)

## AI overview

Dropbox describes using DSPy and evaluation feedback to improve Dash chat. Human-labeled examples calibrate LLM judges, whose scalable signals then optimize the chat agent's system prompt, reducing incomplete answers and token usage without compromising answer quality.

## Source excerpt

We used DSPy to improve LLM judges and optimize our chat experience, creating an evaluation-driven feedback loop that produced better outputs.