Eric Bolme

About Eric Bolme

I work at the intersection of enterprise AI systems, applied machine learning, and product execution, helping organizations turn complex AI capabilities into reliable, production-ready systems—especially during periods of transition, scale, or failure.I’m currently a GenAI Specialist Solutions Architect at AWS, supporting enterprise adoption of Amazon Q Business. My work focuses on diagnosing and stabilizing production GenAI systems, operating enterprise escalation motions, and partnering with product, applied science, and engineering teams to improve retrieval, reasoning, and evaluation behavior in large-scale RAG and agentic systems.Previously, I led enterprise AI initiatives at AWS (Amazon Personalize), DataRobot, McKinsey (QuantumBlack), and Charles Schwab, spanning recommender systems, optimization, fraud and AML, and large-scale ML platforms.I’m typically brought in when AI systems are ambiguous, underperforming, or failing at scale, and teams need to bridge customer needs, scientific rigor, and product execution.Focus areas: Generative AI & RAG • Enterprise AI architecture • Applied ML & evaluation • Recommender systems • Cross-functional technical leadership

Experience

1. Genai Specialist Solutions Architect

   Amazon Web Services (AWS)

   Sep 2024 — Present

   Transitioned into a GenAI Specialist Solutions Architect role to support enterprise adoption of Amazon Q Business during a period of rapid product evolution and customer growth. Operate at the intersection of field, product, and applied science teams, with primary responsibility for enterprise escalations, system diagnosis, and feedback-driven product improvement.Key contributions:Led the Amazon Q Business enterprise escalation motion, initially co-facilitating and later independently running a 3× weekly war room to triage, diagnose, and resolve complex customer issues across retrieval-augmented generation (RAG) and agentic workflows.Coordinated fixes across multiple engineering and applied science teams, resolving systemic issues impacting 70+ enterprise customers.Diagnosed enterprise RAG failures through query-level analysis and led controlled remediation workflows, collaborating with applied science teams to design and evaluate prompt and retrieval strategies (e.g, query rewriting, acronym handling, hybrid retrieval), and partnering with engineering to productionize validated changes.Built synthetic evaluation datasets integrated into existing evaluation pipelines to detect regressions and improve reasoning behavior, with emphasis on conflicting evidence resolution and authoritative source selection.Partnered with product and applied science teams to refine the Amazon Q Business relevancy tuning feature, translating enterprise escalation patterns into customer-driven requirements and authoring evaluation code to execute regression testing and validate ranking improvements.

Education

• New College of Florida

  Bachelor's Degree, History/Mathematics

  2007 — 2011

• University of Connecticut

  Master's Degree, Applied Financial Mathematics

  2013 — 2015

Skills

• Python
• Data Modeling
• Artificial Intelligence
• Pycharm
• Data Analysis
• C#
• Interest Rate Derivatives
• Equity Valuation
• Machine Learning
• Natural Language Processing
• Mathematics Education
• Vb.net
• Networking
• Business Valuation
• Powerpoint
• Microsoft Excel
• Scikit-Learn
• Equity Derivatives
• Desktop Application Development
• Gg2plot
• D3.js
• Process Automation
• Pl/Sql
• R
• Desktop Application Design
• Research
• Matlab
• Visual Studio
• Big Data Analytics
• Front-End Development
• Report Writing
• Quantitative Research
• Numpy
• Gpu Learning
• Tensorflow
• Selenium Webdriver
• Financial Engineering
• Mathematical Modeling
• Microsoft R
• Risk Management