Yuki Mori

Senior ML Engineer focused on Retrieval-Augmented Generation (RAG), NLP, and workflow automation (n8n). Built production-grade data pipelines that combine retrieval, reasoning, and automation to deliver measurable business impact. Deep experience with LLMs, vector databases, hybrid retrieval/reranking, prompt engineering, and end-to-end orchestration on cloud. Looking to drive applied AI systems that improve accuracy, reduce latency, and cut manual ops

Senior AI Engineer - Digital Hero

(2022-10)

Company Description

• Built and deployed production RAG pipelines using LLMs with FAISS/Pinecone and hybrid BM25 + vector + reranking retrieval to improve contextual accuracy
• Orchestrated n8n workflows for ingestion, embedding, retrieval, and inference with retries, backoffs, and alerting for reliable automation
• Delivered NLP capabilities including NER, intent detection, and summarization with iterative prompt and query tuning
• Set up evaluation and observability systems with golden sets, version tracking, and cost/performance dashboards for continuous optimization
• Hardened production through CI/CD pipelines, PII redaction, and runtime guardrails, achieving 99.95% uptime and <0.2% error rate

AI developer - Bilby - Hong Kong

(2019-11 - 2022-05)

• Shipped production Retrieval-Augmented Generation pipelines combining LLMs with FAISS/Pinecone and hybrid BM25 + vector retrieval enhanced by cross-encoder reranking
• Built and maintained n8n automations for ingestion, embeddings, retrieval, and inference with retries, backoff, and alerting for high reliability
• Tuned prompts, retrieval parameters, and query-rewriting logic to strengthen precision, recall, and throughput across large-scale evaluations
• Established robust evaluation and observability frameworks with golden sets, A/B testing, and cost/performance dashboards for continuous optimization
• Implemented production guardrails and compliance controls with CI/CD integration, ensuring 99.95% uptime and <0.20% error rate across services

Internship AI developer - Narrates Limited - Hong Kong

(2018-03 - 2019-03)

• Assisted in developing RAG prototypes that combined LLMs with vector databases (e.g., FAISS/Pinecone) to deliver context-aware internal search and more reliable answers
• Built n8n workflows to automate data ingestion, retrieval, and inference-with retries, backoff, and alerting-to reduce manual steps and accelerate demo readiness
• Tuned prompts and retrieval parameters and executed targeted evaluations (golden sets/spot checks) to improve relevance, stability, and consistency across queries

Bachelor of Science - Computer Science Machine Learning Concentration - Lodz University of Technology (2016-09 - 2019-08)