Chase Bradley

About

ML/LLM engineer from Cohere who has owned retrieval, reranking, agentic workflows, and internal evaluation infrastructure end-to-end in production. Particularly strong in turning brittle RAG and research-heavy ideas into scalable enterprise systems with grounded outputs, lower customer escalations, and adoption by major clients like Notion and Fujitsu.

Key Strengths

• End-to-end ownership of production ML/AI retrieval systems
• Deep expertise in retrieval quality debugging via step-level evals
• Strong RAG and search relevance optimization
• Agentic LLM workflow design for multi-step enterprise use cases
• Grounding and hallucination mitigation through verifier-based gating
• Pragmatic translation of research into scalable production systems
• Latency and cost optimization with caching, pruning, and cascades
• Builds reusable internal platforms that improve team velocity
• Standardized evaluation infrastructure across teams
• Customer-impact focus; reduced escalations on missing obvious results
• Enterprise AI product impact with clients like Notion and Fujitsu

Reference Highlights

• Built Agentic-RAG from 0 to 1
• Set up an evaluation framework used across the entire organization
• Shipped many production RAG systems for business clients
• Improved retrieval quality by upgrading the reranking model
• Expanded enterprise partnerships through technical contributions
• Delivered measurable improvements on client metrics
• Fast shipping
• Strong default in Python
• Useful org-wide code contributions
• Proactive cross-team communicator
• Opened team discussions to resolve tradeoffs and gaps

Experience

Education

Skills

Languages