Nagkumar Arkalgud nagkumar91

## issue_290_trace_showcase.md

      
              1 file
            
          
              0 forks
            
          
                0 comments
              
            
              0 stars
            
          
                nagkumar91
                / issue_290_trace_showcase.md
            
            
              Created
              February 27, 2026 18:36
            
              
                Issue #290 Fix: Detailed Trace Spans & Parent-Child Relationships
              
          
    Issue #290 Fix: Unified Async Agent Trace

Architecture

  Planner (recruitment-planner)
  │
  ├── asyncio.create_task() → worker-rose      (candidate processing + tools)
  ├── asyncio.create_task() → worker-aurora     (candidate processing + tools)
 ├── asyncio.create_task() → eitl-review (interview scheduling + tools)


## issue_290_async_trace_fragmentation.py
"""Reproduce and verify the fix for issue #290: async agent trace fragmentation.

Architecture (from the issue):
  Planner agent dispatches multiple worker agents via asyncio.create_task().
  Workers run concurrently — each has LLM calls and tool use.
  WITHOUT the fix, each worker gets a separate trace_id (fragmentation).
  WITH the fix, all workers share the planner's trace_id (unified trace).

Usage:
    cd libs/azure-ai

## memory-implementation-gen-ai-spec.md

      
              1 file
            
          
              0 forks
            
          
                0 comments
              
            
              0 stars
            
          
                nagkumar91
                / memory-implementation-gen-ai-spec.md
            
            
              Last active
              March 9, 2026 18:42
            
              
                GenAI Memory Operations — Non-Normative Implementation Spec (companion to open-telemetry/semantic-conventions#3250)
              
          
    GenAI Memory Operations — Non-Normative Implementation Spec (companion to open-telemetry/semantic-conventions#3250)
GenAI Memory Operations — Non-Normative Implementation Spec (companion to open-telemetry/semantic-conventions#3250)
GenAI Memory Operations — Non-Normative Implementation Spec (companion to open-telemetry/semantic-conventions#3250)
Memory Operations for GenAI (Revised Implementation Proposal)

Why Memory Matters for Observability:
	"""Reproduce and verify the fix for issue #290: async agent trace fragmentation.

	Architecture (from the issue):
	Planner agent dispatches multiple worker agents via asyncio.create_task().
	Workers run concurrently — each has LLM calls and tool use.
	WITHOUT the fix, each worker gets a separate trace_id (fragmentation).
	WITH the fix, all workers share the planner's trace_id (unified trace).

	Usage:
	cd libs/azure-ai