Remote Agentic AI/LLM Engineer, Coverage Intelligence

Bold Penguin, Inc.; Dublin (OH); Remote; USD 140,000 - 200,000; Full time

Benefits offered by this job

Annual Performance Bonus Eligible

Medical, Dental, and Vision

Flexible PTO Policy

401(k) with company match

Employee Assistance Program

Disability and Life Benefits

Job summary

Bold Penguin, Inc. is seeking an Agentic AI / LLM Engineer to design and build AI systems for converting complex insurance documents into reliable coverage intelligence. You will own end-to-end problems from understanding unstructured documents to building agents that reason about coverage.

You will ground outputs in source documents, collaborate with domain experts and engineers, and drive accuracy through rigorous evaluation and iterative improvement.

Qualifications

• 5+ years' experience in software engineering or applied machine learning.
• 2+ years building LLM applications in production.
• Strong Python skills with typed schemas (e.g., Pydantic).
• Hands-on experience extracting structured data from complex documents (PDF parsing, OCR).
• Experience with agentic patterns: tool use, multi-step workflows, human-in-the-loop.

Responsibilities

• Design, build and own production AI systems for document understanding and coverage analysis.
• Build agentic workflows combining models, tools and human review.
• Ground outputs in source documents and domain knowledge for traceability.
• Drive accuracy via evaluation, experimentation and error analysis.

Skills

LLM applications in production

Python

PDF parsing/ OCR

Agentic workflows

Retrieval-augmented generation

Evaluation/metrics

Communication

Autonomy in work

LangGraph

LLM observability tooling

Tools

LangGraph

LLM observability tooling

Job description

Bold Penguin, Inc. is seeking an Agentic AI / LLM Engineer to design and build AI systems for converting complex insurance documents into reliable coverage intelligence. You will own end-to-end problems from understanding unstructured documents to building agents that reason about coverage.

You will ground outputs in source documents, collaborate with domain experts and engineers, and drive accuracy through rigorous evaluation and iterative improvement.