AWS Certified Solutions Architect Β· AI Systems Β· Amsterdam, NL π³π±
I build production-grade AI and cloud infrastructure β from serverless pipelines to multi-agent systems. Currently transitioning into Solutions Architect and Solutions Engineer roles at AI and cloud companies.
- π‘ AWS Certified AI Practitioner
- π AWS Certified Solutions Architect β Associate
- β¬ AWS Certified Cloud Practitioner
Production-grade RAG system enabling lawyers to ask natural language questions across an entire contract library and receive precise answers with verifiable source citations. Includes multilingual support, automated ingestion, and pre-signed URL document access.
Stack: Bedrock Knowledge Bases Β· OpenSearch Serverless Β· Lambda Β· API Gateway Β· S3 Β· Terraform IaC
π 99.8% cost reduction vs manual associate review Β·
Local RAG pipeline built to understand what managed services abstract away β vectors, chunking, retrieval scoring, and prompt construction β instrumented end-to-end with LangSmith tracing and evaluation.
The previous project proved I can build a RAG pipeline. This one proves I understand what's happening inside it.
Stack: LangChain Β· Chroma Β· Voyage AI Β· Claude Haiku Β· LangSmith
π Full eval suite with Claude-as-judge scoring Β· Known failure modes documented with proposed fixes Β· View Project
Event-driven pipeline automating insurance claims processing end-to-end.
Stack: S3 β Lambda β Textract β Bedrock β DynamoDB β SNS/SQS Β· Terraform IaC
π 98.7% cost reduction vs manual processing Β·
Production-pattern HA infrastructure β ALB, ASG, EC2, RDS Multi-AZ, dual NAT Gateways.
Stack: Terraform Β· AWS VPC Β· RDS Β· EC2
- π’ Solutions Architect @ Latent Knowledge
- π’ Transitioning into a Forward Deployed Engineer role β embedding AI capabilities within business functions at Just Eat Takeaway
- π οΈ Building technical depth in LLM observability, evaluation, and retrieval systems