Public Sector Healthcare Out-of-Pocket Cost Optimizer with Google ADK and Together AI
Rising healthcare costs strain public sector budgets and employee benefits. You will build a Python agent using Google ADK and Together AI inference to process claims datasets, optimize benefit plan coverage rules, and reduce out-of-pocket expenditure predictions by at least 15% across employee cohorts.
What you are building
The core problem, expected build, and operating context for this challenge.
Construct a Google Agent Development Kit pipeline with Together AI inference to optimize municipal health insurance claim structures.
How work is evaluated
Evaluates cost-saving recommendation accuracy and model execution efficiency on municipal claims data.
Challenge datasets
Public datasets, plus the private datasets you have access to.
How submissions are scored
These dimensions define what the evaluator checks and which criteria separate a passable run from a strong one.
positive_savings_check
Ensure projected savings exceed 15% minimum threshold
This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.
savings_percentage
Percentage reduction in out-of-pocket costs • target: 18 • range: 0-100
This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.
What you should walk away with
Implement multi-agent workflows using Google ADK (Agent Development Kit)
Connect Together AI model endpoints for high-throughput claim optimization
Analyze out-of-pocket cost distribution across diverse public employee cohorts
Reference links and supporting material
Municipal employee healthcare claims dataset comprising 10,000 anonymized annual claim records.
Challenge setup
Evaluates cost-saving recommendation accuracy and model execution efficiency on municipal claims data.
Challenge input
JSON array of annual claim records
Google ADK
Mandatory Google agent development SDK.
Together AI
Provides fast open-source model execution.
Evaluated output
JSON object with recommended copay limits and predicted savings float
- Ensure projected savings exceed 15% minimum threshold
- Percentage reduction in out-of-pocket costs • target: 18 • range: 0-100
- Benchmark: HealthEcon-Bench
- Savings Percentage target: 18
- 1 public reference case
- Python execution harness
- Python sandbox (unavailable on Versalist)
Summary shows the first task, up to three tools, and up to three evidence items. Open the technical recipe to view all configured items.
View technical recipe
Configured tools
- Google · Required
- Google ADK · Optional
- Together AI · Optional
- Together AI · Optional
Evaluation contract
- positive_savings_check · Weight 1
- savings_percentage · Weight 1
Metrics
- savings_percentage · Target 18
Benchmarks
- HealthEcon-Bench
- general
Reference cases
- 1 public reference case
Workflow tasks
- Task 1
Input: JSON array of annual claim records
Output: JSON object with recommended copay limits and predicted savings float
Execution harness: Python
Evaluation tools
- Google ADK · Agent Framework
- Together AI · Inference Engine
Recipe freeze marker
No recipe freeze time is recorded.