back_office_ops · travel · workflow
JetBlue uses DSPy and Databricks to build self-optimizing multi-tool LLM pipelines
Multi-stage LLM pipelines required continuous manual prompt tuning, where single words could make or break a deployment, creating brittle and time-consuming iterative development.
How it works
Common implementation structure
How this type of workflow is generally built, generalized across documented cases — not tied to any one vendor's stack. Click any stage to read what happens there. Specific products that implement these stages appear in “Tools commonly seen” below.
Stage 1 · User question submitted
A user question enters the pipeline as the initial input.
Tools used
DSPyDatabricksDatabricks Model ServingDatabricks Vector SearchMLflowLangchainDBRXLlama 2 70BOpenAI
Outcome
JetBlue deployed its RAG chatbot 2x faster than with Langchain, eliminated the need to manually iterate on prompts, and launched revenue-driving customer feedback classification and RAG-powered predictive maintenance chatbots.
What failed first
Before DSPy, JetBlue relied on manually crafted, highly explicit prompt templates requiring precise hand-written instructions for each pipeline step.
Results
Volume2x faster
Grounding & classification
Source type: technical build writeup
29 fields verified against source quotes.
agentic workflowchatbotdocument classificationragknowledge basefailure mode describedmetric backednamed customerproduction runtime claimedtools describedworkflow describedtravelcycle time reductionemployee productivitytechnical build writeupback office opsagentic task executionrag answering