Back office ops · Production

Nayya advances personalized benefits recommendations using Labelbox for AI data labeling and model evaluation

The problem

Nayya needed faster and cheaper ways to produce labeled training data for ML model development, live prediction evaluation, and subject matter expert verification, drawing on diverse data sources including insurance plans, pharmaceutical data, geospatial data, and financial wellness insights.

Workflow diagram · grounded in source
1
User survey submission
trigger
“Users take a ten-minute survey and get a plan recommendation that best suits their specific needs”
2
ML model recommendation
ai_action
“Machine learning models work behind the scenes to power decision support tools for a personalized experience that includes financial education and bundled recommendations”
3
Labelbox annotation workflow
integration
“The annotation workflow within Labelbox allows us to do this at scale and build a repeatable process for our data scientists as well as any subject matter experts that work with us”
4
SDK-driven SME prediction review
human_review
“an automated, SDK-driven approach to allow their actuaries (SMEs) to share insights into how the model will generate predictions and evaluate these predictions”
5
QA and data cataloging
validation
“the team has set up a strong QA process within the platform to help speed up their efforts when it comes to cataloging and extracting the data from disparate sources”
6
Model recommendation feedback loop
feedback_loop
“the organization uses an automated feedback loop between their customer success team and benefit providers to evaluate and explain model-generated recommendations”
Reported outcome

Nayya's ML team can now better visualize unstructured data, maintain training data quality, save time, and train and test new models faster through more streamlined labeling and enhanced collaboration, with reproducible MLOps pipelines in place.

Reported metrics
time savings for ML teamsave time
Model training speedtrain models faster
Training data capacitykeep up with their fast-growing training data needs
Training data qualitymaintain training data quality
Reported stack
LabelboxLabelbox AnnotateCatalogPython SDKfeature store
Source
https://labelbox.com/customers/nayya-customer-story/
Read source ↗

Frequently asked questions

What did this team achieve with this AI workflow?

Nayya's ML team can now better visualize unstructured data, maintain training data quality, save time, and train and test new models faster through more streamlined labeling and enhanced collaboration, with reproducib…

What tools did this team use?

Labelbox, Labelbox Annotate, Catalog, Python SDK, feature store.

What results were reported?

time savings for ML team: save time; Model training speed: train models faster; Training data capacity: keep up with their fast-growing training data needs; Training data quality: maintain training data quality (source-reported, not independently verified).

How is this back office ops AI workflow structured?

User survey submission → ML model recommendation → Labelbox annotation workflow → SDK-driven SME prediction review → QA and data cataloging → Model recommendation feedback loop.

Related back office ops cases