Purdue Model Multi-Tier Telemetry Bridge with LangChain and LangGraph
Scaling Industry 4.0 applications requires strict adherence to security boundaries across Purdue Model Levels 0-4 without introducing high-latency choke points. You will build a multi-tier data transformation and routing pipeline using LangChain and LangGraph to securely aggregate OT edge metrics into Enterprise Level 4 databases. The solution must enforce strict Purdue level segregation while keeping stream mapping latency under 100ms.
What you are building
The core problem, expected build, and operating context for this challenge.
Construct a LangGraph stateful graph agent that securely bridges process data across Purdue Model levels while validating network security boundaries.
How work is evaluated
Evaluates Purdue isolation boundary checks and data conversion correctness.
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.
ip_sanitization_check
Verifies lower-level OT IP addresses are stripped out before reaching enterprise output.
This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.
conversion_latency_ms
Latency through full Purdue graph pipeline • target: 85 • range: 0-500
This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.
What you should walk away with
Design stateful multi-tier workflow graphs using LangGraph
Implement security dynamic inspection logic preventing direct Level 0/1 sensor access from Level 4 cloud queries
Use LangChain tools to parse and re-map OPC UA / Modbus payloads into enterprise IT JSON schemas
Measure data isolation integrity and benchmark bridge processing latency
Reference links and supporting material
Multi-tier factory network logs containing Modbus register dumps, SCADA alarms, and MES work order events.
Challenge setup
Evaluates Purdue isolation boundary checks and data conversion correctness.
Challenge input
JSON containing raw_tag_data, source_purdue_level, and target_purdue_level
LangChain
Required framework assignment.
LangGraph
Required framework sub-assignment.
Evaluated output
JSON with mapped_enterprise_payload, dynamic_sanitization_passed, and path_latency_ms
- Verifies lower-level OT IP addresses are stripped out before reaching enterprise output.
- Latency through full Purdue graph pipeline • target: 85 • range: 0-500
- Benchmark: Purdue Boundary Integrity Benchmark
- Conversion Latency Ms target: 85
- 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
- LangGraph · Required
- Langchain · Optional
- LangChain · Optional
- LangGraph · Required
- Langchain · Optional
- LangChain · Optional
Evaluation contract
- ip_sanitization_check · Weight 1
- conversion_latency_ms · Weight 1
Metrics
- conversion_latency_ms · Target 85
Benchmarks
- Purdue Boundary Integrity Benchmark
- security
Reference cases
- 1 public reference case
Workflow tasks
- Task 1
Input: JSON containing raw_tag_data, source_purdue_level, and target_purdue_level
Output: JSON with mapped_enterprise_payload, dynamic_sanitization_passed, and path_latency_ms
Execution harness: Python
Evaluation tools
- LangChain · Agent Framework Base
- LangGraph · Stateful Orchestrator
Recipe freeze marker
No recipe freeze time is recorded.