About the Role
Pipeline Monitoring & Triage: activelymonitor live Agentic workflows and scraping pipelines. Identify failures (e.g., stuck agents, failed scrapes, API timeouts), diagnose the root cause, and determine if it requires a quick fix or an escalation.
L1/L2 Code Maintenance: Perform initial debugging of Python-based automation scripts. Fix minor logic bugs, dependency issues, or data formatting errors in the codebase.
Prompt Engineering & Adjustment: Analyze LLM outputs for quality. Perform minor adjustments to system prompts or parameters (temperature, max tokens) to correct formatting or adherence issues without retraining models.
RAG Operations: Assist in maintaining vector databases (Pinecone/Milvus). Handle re-indexing requests and verify that semantic search results are retrieving relevant context (checking for data staleness or embedding mismatches).
AWS Operational Support: Manage the...