Build a Waste-to-Energy Process Optimization Agent with Mastra AI and Agency
Waste-to-energy facilities require precise thermal conversion balancing and emissions monitoring to maintain municipal compliance. Build a workflow automation agent using Mastra AI and Agency observability to parse waste composition logs and optimize burner setpoints, achieving at least 95% throughput accuracy under peak load.
What you are building
The core problem, expected build, and operating context for this challenge.
Implement an automated waste-to-energy tuning agent using Mastra AI workflows integrated with Agency telemetry tools.
How work is evaluated
Evaluates Mastra AI agent accuracy and Agency trace logging on waste-to-energy telemetry inputs.
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.
temperature_safety_check
Ensures combustion chamber temperature stays between 850C and 1100C
This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.
setpoint_accuracy
Proportion of calculated air-fuel ratios within target thermal efficiency bounds • target: 0.95 • range: 0-1
This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.
What you should walk away with
Construct step-based workflow graphs using Mastra AI in TypeScript
Integrate Agency for complete observability and agent execution tracking
Parse unstructured municipal solid waste composition reports and sensor feeds
Evaluate thermal setpoint recommendations against environmental safety thresholds
Reference links and supporting material
Synthetic telemetry dataset representing Casablanca municipal solid waste streams, caloric density, and combustion output metrics.
Challenge setup
Evaluates Mastra AI agent accuracy and Agency trace logging on waste-to-energy telemetry inputs.
Challenge input
JSON containing moisture_percentage, caloric_value_mj_kg, throughput_tons_hr
Mastra AI
TypeScript-native agent workflow orchestration
Agency
Execution tracking and telemetry logging
Evaluated output
JSON containing primary_air_flow_m3_h, secondary_air_flow_m3_h, expected_emissions_nox_ppm
- Ensures combustion chamber temperature stays between 850C and 1100C
- Proportion of calculated air-fuel ratios within target thermal efficiency bounds • target: 0.95 • range: 0-1
- Benchmark: Industrial AI Process Benchmark
- Setpoint Accuracy target: 0.95
- 1 public reference case
- JavaScript execution harness
- javascript 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
- Mastra AI · Required
- Agency (Observability) · Optional
- Bland AI · Optional
- Mastra AI · Required
Evaluation contract
- temperature_safety_check · Weight 1
- setpoint_accuracy · Weight 1
Metrics
- setpoint_accuracy · Target 0.95
Benchmarks
- Industrial AI Process Benchmark
- energy
Reference cases
- 1 public reference case
Workflow tasks
- Task 1
Input: JSON containing moisture_percentage, caloric_value_mj_kg, throughput_tons_hr
Output: JSON containing primary_air_flow_m3_h, secondary_air_flow_m3_h, expected_emissions_nox_ppm
Execution harness: JavaScript
Evaluation tools
- Mastra AI · Agent Framework
- Agency · Observability
Recipe freeze marker
No recipe freeze time is recorded.