Wondr Health scales trusted health coaching with Claude
Wondr HealthClaude Platform · Amazon Bedrock · LangGraph +2Wondr Health, a digital behavior-based weight management program offered as an employer benefit, worked with AI engineering firm Blank Metal to build Wonda, an AI coaching layer running on Claude via Amazon Bedrock: Claude Opus 4.5 powers coaching conversation reasoning, Sonnet runs a parallel safety subagent that monitors every conversation turn and escalates risk to human coaches via Zendesk, and Haiku runs automated evaluations, with onboarding answers grounded in Wondr's curriculum through a RAG pipeline. Before launch, Wondr's clinical and coaching teams tested Wonda against more than 700 questions across six user personas. The AI coaching layer compresses onboarding that historically took a human coach 30 to 60 minutes into a 10-minute guided conversation.
Overview
Wondr Health, a digital behavior-based weight management program offered as an employer benefit, worked with AI engineering firm Blank Metal to build Wonda, an AI coaching layer running on Claude via Amazon Bedrock: Claude Opus 4.5 powers coaching conversation reasoning, Sonnet runs a parallel safety subagent that monitors every conversation turn and escalates risk to human coaches via Zendesk, and Haiku runs automated evaluations, with onboarding answers grounded in Wondr's curriculum through a RAG pipeline. Before launch, Wondr's clinical and coaching teams tested Wonda against more than 700 questions across six user personas. The AI coaching layer compresses onboarding that historically took a human coach 30 to 60 minutes into a 10-minute guided conversation.
This entry has 15 published fields tied to exact passages in an immutable source capture.
Inspect the highlighted sourceThe challenge
Wondr wanted to scale its coaching programs to serve more people while making coaching relationships stronger, starting with onboarding, where the trust-building that historically took a human coach 30 to 60 minutes had to happen in a 10-minute digital window without losing participants who don't find their footing early. Coaches were stretched thin fielding high-volume, repetitive questions while needing to analyze data across systems to understand a single participant's situation.
The solution
Wondr partnered with AI-native engineering firm Blank Metal across a three-sprint engagement to build Wonda, an AI coaching layer powered by Claude through Amazon Bedrock. Claude Opus 4.5 powers the coaching agent's reasoning, Claude Sonnet runs a parallel safety subagent that monitors every conversation turn and escalates risk to human coaches via a Zendesk ticket, and Claude Haiku runs automated evaluations as an LLM judge, with onboarding answers grounded in Wondr's curriculum through a retrieval augmented generation (RAG) pipeline. LangGraph handles orchestration and LangSmith captures full conversation tracing, and a Next.js testing console let clinical and coaching teams put 700+ questions through Wonda across six user personas without engineering support.
Reported business value
Wonda delivers onboarding in a 10-minute conversation, compressing what historically took a coach 30 to 60 minutes and letting participants get to human coaching more efficiently. Between the two human evaluation rounds, Wonda's scored response quality kept improving, with strong reception on tone, empathy, and personalization from the coaching and clinical teams, and every repetitive onboarding question Wonda takes on is meant to free a coach for conversations that require a human.
Sources
Open any source and check the claim yourself — that is the point of the register.
This record was researched and written with AI assistance, and its claims were checked against the sources above. (EU AI Act art. 50 transparency notice.)
Other technology & software entries in the register.
HP crafts marketing campaigns that resonate with customers using Databricks and Uniphore
HP centralized first-party customer data on the Databricks Data + AI Platform with Delta Lake and Unity Catalog, and connected it to Uniphore's HybridCompute for federated query pushdown, cutting campaign setup from 2 weeks to 2 hours and processing 400 million records in seconds.
Building a safer and more sustainable world
Novade, a construction management software company, partnered with Databricks to modernize its analytics and ML infrastructure, moving from a homegrown Apache Airflow stack to the Databricks Data + AI Platform with Delta Lake, Unity Catalog and MLflow. The move reduced total cost of ownership by 60% and supported a 100% increase in new clients, while enabling ML models for predicting incident risk affecting worker safety and project delivery schedules.
Transforming Weather Forecasting with Lakeflow Jobs
AccuWeather migrated from on-premises infrastructure to Databricks and Lakeflow Jobs, working with Datadog for observability, to unify diverse weather data formats and orchestrate 4,500+ weekly jobs. Lakeflow Jobs coordinates the ingestion of multiple weather models, triggers machine learning processes that weight and blend different forecasts, and manages complex job dependencies for reinforcement training workflows used in AccuWeather's proprietary forecasting engine. AccuWeather reports 3x faster dataset development (three months to one month per dataset), a 50% reduction in unactionable alerts, and 50% cost savings on serverless job usage.
AccuWeatherLakeflow Jobs · Delta Sharing · DatadogAdobe brings creativity to life with Databricks
Adobe uses the Databricks Data + AI Platform for end-to-end data management that unifies all data and AI at scale, with 20% faster performance. Databricks equips over 92 teams at Adobe to unify data from financials, sales, products, customers and employees so they can drive personalized experiences across Adobe's digital platforms with AI.
Was this helpful?
Your feedback helps us improve our use case database