Hari Krishna Kona

About

LLM/AI engineer who built a production automated document-understanding pipeline on Azure using a grounded RAG layer, designed to reduce manual review time for unstructured financial documents. Demonstrates strong real-world scaling and reliability practices (Service Bus queueing, Kubernetes autoscaling, observability, retries/circuit breakers) plus rigorous evaluation (shadow testing, replaying traffic, multilingual edge-case suites) and stakeholder-friendly, evidence-based explainability.

Key Strengths

• Built and deployed production LLM document-understanding pipeline with RAG grounding on Azure
• Scaled LLM processing for long documents via chunking + queueing (Service Bus) + Kubernetes workers
• Reliability engineering for LLM services (structured logs/metrics, retries with exponential backoff, circuit breakers)
• Strong orchestration experience across batch and real-time workloads (ADF, Airflow DAGs/sensors, Kubernetes autoscaling)
• Test discipline for data/ML pipelines (unit tests for operators/activities; staging runs with realistic volumes)
• Practical LLM evaluation approach (replay/shadow traffic, edge-case suites, multilingual and mixed-language failure patterns)
• Business-aligned model/prompt selection focused on worst-case behavior and cost of errors, not just average accuracy
• Effective communication with non-technical stakeholders using evidence-based outputs to build trust

Experience

Education

Skills

Languages