← Back to Cutting Onboarding From 28 Days to Hours With GenAI

Source proof for Cutting Onboarding From 28 Days to Hours With GenAI

Source-bound proof

Verified source excerpts for every supported field

Each colour maps a published value to the exact source passage used to support it. Only bounded excerpts are public; administrators can inspect the complete captured source.

11 fields supported

Title

Cutting Onboarding From 28 Days to Hours With GenAI

derived · high
…a Demo Login Contact Us Try Databricks Customer Stories / Mirakl CUSTOMER STORY Cutting onboarding from 28 days to hours with genAI 0 Ops overhead equal to 5 full-time engineers 24 Hours for supplier catalog onb…

Company

Mirakl

classification · high
…rity and Trust Ready to get started? Get a Demo Login Contact Us Try Databricks Customer Stories / Mirakl CUSTOMER STORY Cutting onboarding from 28 days to hours with genAI 0 Ops overhe…

Description

Mirakl, an eCommerce software platform supporting over 100,000 merchants, used the Databricks Data + AI Platform to build Catalog Transformer, a GenAI solution that automates vendor catalog onboarding. Using Delta Lake, Unity Catalog, MLflow, Lakeflow Jobs and Model Serving running GPT, Llama, Mistral and CLIP models, Mirakl cut supplier catalog onboarding from an average of 28 days to under 24 hours (a 91% reduction) with roughly 50% fewer categorization errors, and extended the platform to Mirakl Nexus for agentic commerce.

derived · high
…erating activities. Together with OpenAI models running natively on Databricks, Catalog Transformer achieves a 91% reduction in onboarding time and approximately 50% fewer categorization errors, directly improving time-to-revenue for merchants and the shopping experience f…

Problem

This process traditionally took an average of 28 days for each new vendor catalog to be processed. These constraints became especially limiting as Mirakl began exploring large language models for automating supplier catalog onboarding.

derived · high
…loring large language models (LLMs) for automating supplier catalog onboarding. This process traditionally took an average of 28 days for each new vendor catalog to be processed. To keep pace with merchant demand and unlock GenAI at scale, Mirakl needed a si…

Implementation approach

A multi-stage pipeline, orchestrated with Lakeflow Jobs, serves endpoints through Model Serving, which runs a mix of models — including GPT, Llama, Mistral, and multimodal models like CLIP — to categorize, extract, enrich, and automatically rewrite vendor data.

derived · high
…and still know exactly which model version and dataset powered each decision.” A multi-stage pipeline, orchestrated with Lakeflow Jobs , serves endpoints through Model Serving , which runs a mix of models — including GPT, Llama, Mistral, and multimodal mo…

Business value

Catalog Transformer achieves a 91% reduction in onboarding time and approximately 50% fewer categorization errors. A senior data engineer estimates they would need five additional engineers to maintain equivalent infrastructure on their own.

derived · high
…erating activities. Together with OpenAI models running natively on Databricks, Catalog Transformer achieves a 91% reduction in onboarding time and approximately 50% fewer categorization errors, directly improving time-to-revenue for merchants and the shopping experience f…

Industry

Technology & Software

classification · high
…ing both internal productivity and customer experience. Share this post Details Industry : Technology and Software Use Case : Artificial Intelligence , Data Applications , Governance and Securit…

Technology

Delta Lake, Unity Catalog, MLflow, Lakeflow Jobs, Model Serving, Agent Bricks

classification · high
…ison, multi-merchant baskets, transaction execution, and post-purchase support. Mirakl Nexus uses Agent Bricks to orchestrate production-ready agents that can safely reason over retailer, merchant, and catalog data. Unity Catalog…

Business functions

Supply Chain & Logistics

classification · medium
…especially limiting as Mirakl began exploring large language models (LLMs) for automating supplier catalog onboarding. This process traditionally took an average of 28 days for each new vendor cata…

AI capabilities

Generative AI, Agentic AI, Computer Vision

classification · high
…Model Serving , which runs a mix of models — including GPT, Llama, Mistral, and multimodal models like CLIP — to categorize, extract, enrich, and automatically rewrite vendor data. Spark…

Deployment options

cloud

classification · high
…Intelligence , Data Applications , Governance and Security , Data + AI Platform Cloud : AWS Product : Agent Bricks , Lakeflow Jobs , Unity Catalog Ready to get started? Tr…
Capture details
Captured
04 Sept 2026, 06:03 UTC
Extractor
fetch-strip@1
Snapshot hash
eeb6ef222f2211d82ec1166ab7e85af8a7c6611d8a36fb48df61d353fba97eae