Anis Khai

Scaled document AI and GenAI pipelines to triple throughput; reduced analyst time by ~40% using RAG-based automation and semantic retrieval. Standardized enterprise GenAI reference architecture adopted across 15+ squads. Led end-to-end RAG for enterprise knowledge: ingestion, parsing, chunking, embeddings, indexing, hybrid retrieval, reranking, and grounding with citations. Architected LLM-based assistants and agentic workflows; optimized GPU-accelerated serving on Kubernetes; established governance and CI/CD/IaC practices; embedded access control and data classification for sensitive content.

Delivered Kubernetes reference architectures across on-prem and Azure; implemented security baselines; designed high-availability platforms; produced HLD/LLD for delivery teams. Implemented backup/restore strategies; integrated observability foundations (OpenTelemetry/Prometheus) and alerting. Strengthened security with image scanning and identity integration; collaborated with stakeholders and agile delivery teams.

Standardized Infrastructure-as-Code (Terraform) and deployment patterns; improved reliability and observability for microservices platforms. Designed cloud architectures on Azure and AWS focusing on availability, performance, security, and operational excellence. Translated architecture into Terraform modules and runbooks; supported deployments and release readiness; implemented centralized logging/monitoring (Elasticsearch/Kibana) and performance testing.