Evaluate how well models detect agent disconnections, stale state, and functional failures from labeled endpoint telemetry time series.

Train models to acknowledge evidence gaps, avoid fabrication, and ask useful follow-up questions with multi-turn examples built around insufficient evidence.

Connect process events with orders, invoices, and deliveries to build knowledge graphs that represent cross-object process relations.

Find notification and escalation-loop risks with verifiable labels across diverse on-call alert scenarios.

Train models to trace how resource state shapes distributed job placement, from scheduler code to placement decisions.

Train traceable 5G SA reasoning with consistency checks across subscriptions, session requirements, and network slicing policies.

Help coding agents understand how cloud-native instrumentation generates telemetry and propagates context.

Trace transitive dependencies from project manifests and train models to assess version compatibility and pinpoint conflicts.

Train models to generate scene structures, component configurations, and initialization code from natural-language requests and paired engine code.

Train coding agents to identify unauthorized access and tenant-isolation defects with enterprise chat authorization code, tests, and repair examples.

Train models to generate and refine streaming SQL through interactive requirement clarification and query iteration.

Pair source code with authorization failure analysis and fixed examples to strengthen model reasoning about microservice security boundaries.

Analyze exactly-once semantics in distributed event processing through source code, dependencies, and the impact of code changes.

Train interpretable fault diagnosis with 5G core session establishment cases, linked evidence, and practical validation steps.

Train models to identify concurrency mechanisms, shared state, and potential race conditions in distributed cache code through source-based question answering.

Evaluate code verification and root-cause analysis with 5G core protocol codec and information element edge cases.

Train coding agents to analyze messaging platform security code, trace authorization paths, locate flaws, and plan fixes.

Evaluate scheduler leader election and failover logic with code samples, failure scenarios, and root cause evidence.

Trace how SIP authentication and tenant authorization controls affect call setup, routing, and processing.

Evaluate how well models detect defects and trace root causes in travel API mappings using supplier response samples and validation cases.