Eric Voss
I build, evaluate and optimize responsible AI solutions for regulated industries.
I'm a Minnesota-based data scientist at Securian Financial, where I build agentic AI systems and help set enterprise standards for AI observability and evaluation, all in an industry where PII and PHI handling, HIPAA, and model governance are critical. My career spans more than a decade across application development, analytics, and AI/ML in the financial services industry.
I have extensive experience in developing AI solutions with the AWS ecosystem, leveraging platforms like Bedrock and AgentCore. My previous experience also includes "traditional" data analysis and predictive modeling.
What I work on
More about me
Intelligent document processing
Designing and building AI pipelines that turn messy, low-quality documents into structured, enriched data for consumption by human and AI users.
Agentic AI architecture
Architecting and building agentic systems on AWS, primarily Amazon Bedrock AgentCore — tool and memory design, orchestration (both deterministic and agentic), and the necessary guardrails an agent needs before it is allowed near sensitive workflows.
Evaluation & enterprise AI standards
Systematic evaluation of AI systems: error taxonomies, test sets, and regression checks — and the standards and practices that define how AI gets built, reviewed, and operated across the company.