- 10k+ PDFs RAG Pipeline: Built an end-to-end ingestion and vectorization pipeline, tuning retrieval quality (via LangFuse tracing) to achieve a +22pt improvement in search success.
- 60% Faster Deployments: Architected secure, automated delivery with GitHub Actions, Docker, Trivy, and Cosign, slashing deployment time from 18 to 7 minutes.
- Certified Expertise: AWS Certified Solutions Architect (Professional), OCI AI Foundations (Associate), and OCI Foundations (Associate).
|
Designing practical RAG workflows with AWS Bedrock, Dify, OpenSearch, Python, and Langfuse. Evaluating model responses and optimizing vector search retrieval. |
Building repeatable delivery pipelines with GitHub Actions, Docker, GHCR, Trivy, Cosign, SBOMs, canary releases, and rollback-aware release flows. |
|
Implementing Datadog metrics, logs, APM, monitors, and LLM observability (LangFuse). Experience with GKE production monitoring and infrastructure-as-code alerts. |
Operating public services and private lab workloads across AWS, Azure, GCP, Cloudflare, Vercel, and self-hosted infrastructure with Zero Trust boundaries. |