Senior Applied AI Software Engineer
Senior Applied AI Engineer building LLM-powered applications for MediaAlpha's internal teams and ad serving platform, owning features end-to-end from integration through production deployment. Core work includes MCP servers, RAG pipelines, vector search, agent workflows, and prompt engineering on AWS.
- Compensation
- $150K–$210K / yr
- Location
- Bellevue, WA
- Work setup
- Hybrid
- Seniority
- Senior
- Employment
- Not specified
- Experience
- 5+ years
- Category
- Ai engineering
- Posted
Skills
Must have
- Minimum of 5 years of experience in software development
- B.S. required, preferably in a technical discipline (e.g., computer science)
- Strong programming skills, including software engineering methodologies and best practices
- Strong SQL skills and understanding of database and query optimizations
- Hands-on experience building LLM-powered features in production
- Familiarity with common AI implementation patterns: MCP server design, RAG, vector search, function calling/tool use, agent orchestration, structured output extraction
- Practical prompt engineering skills with demonstrated ability to iterate and evaluate quality
- Experience integrating with APIs and building services that call external or internal LLM endpoints
- Familiarity with cloud environments (AWS preferred: EC2, S3, Bedrock)
- Ability to work independently with minimal supervision
- Self-starter, comfortable with ambiguity, careful attention to detail
- Experience working with engineers and non-technical stakeholders
- Ability to articulate and communicate clearly, concisely, and thoughtfully
Full description
MediaAlpha is a customer acquisition solutions provider powered by technology and data science. The company provides industry-leading solutions designed to reach consumers shopping within high-consideration categories such as property and casualty insurance, health insurance, life insurance, and more.
Hybrid role: 3 days/week in office
Role Overview
As a Senior Applied AI Engineer at MediaAlpha, you will be a key contributor building LLM-powered applications for our internal teams and ad serving platform. You will implement features end-to-end: integration, deployment, and operational health. You will work alongside engineering, product, and business stakeholders to deliver features that solve real business problems.
The ideal candidate is a strong software engineer with hands-on experience building LLM-powered features in production, who works well across organizational boundaries and improves the conditions of any area they are in. We are an intentionally small organization, designed for agility, speed, and workability.
Responsibilities
- Own the full software development lifecycle of implementing LLM-powered features, from integration through production deployment and operational health.
- Design and build MCP (Model Context Protocol) servers that expose internal data and tools to LLM applications.
- Design, implement, and maintain retrieval-augmented generation (RAG) pipelines, including vector search, tool-use patterns, and multi-step agent workflows.
- Integrate LLM-based services into MediaAlpha's products, internal tools, and data ecosystem.
- Write and iterate on prompts, and build the harnesses needed to test and evaluate them.
- Be part of the team that maintains and supports our live production AI features.
- Be a champion of instilling excellence in coding, testing, and quality into the team and culture.
- Work directly with stakeholders in defining product capabilities and developing and testing new features and integrations.
Requirements
- Minimum of 5 years of experience in software development.
- B.S. required, preferably in a technical discipline (e.g., computer science).
- Strong programming skills, including software engineering methodologies and best practices.
- Strong SQL skills and understanding of database and query optimizations.
- Hands-on experience building LLM-powered features in production.
- Familiarity with common AI implementation patterns: MCP server design, RAG, vector search, function calling/tool use, agent orchestration, and structured output extraction.