AI Engineer

o Frontend/backend for RAG and app work: Next.js and NestJS (Node) o Light work with data pipelines: Python; Snowflake as the data platform (medallion architecture: bronze/silver/gold)

• Tools and AI coding assistants:

o Claude Code o GitHub Copilot via Visual Studio o Evaluating vendor AI tools (e.g., Snowflake AI, Domo AI); use-case dependent. o MCP servers: discussed; on roadmap; not currently required for internal LLM routing/abstraction.

Do you have experience in Software engineering?, * Strong Python experience (production-grade software engineering).

• Hands-on experience working with LLMs in production (general LLM best practices; not strictly RAG). Examples:

o Efficient interaction patterns with LLMs (token management, sending full articles vs. selective context) o Agentic approaches for complex reasoning (e.g., applying AP style guide across thousands of rules) o Practical strategies to avoid context overload and maintain relevance.

• Ability to "run with projects," operate independently, and collaborate with stakeholders.

• Minimum experience: approximately 5 years; must have "done it before."

Should Have

• Familiarity with Next.js/NestJS/Node for application/RAG-related work; strong Python candidates can ramp with AI coding tools.

• CI/CD experience; Terraform not required (team strength exists, can learn on the job).

• Good culture fit: collaborative, mission-driven, able to navigate flexible stack choices aligned with product teams.

Could Have:

• Exposure to data engineering concepts and tooling:

o Building ingestion/ETL/ELT pipelines (Python) o Working with Snowflake; experience in similar platforms (Redshift, Synapse) acceptable with ability to translate principles. o Familiarity with medallion architecture and data modeling concepts is helpful but not strictly required (team can support ramp-up).

$60 - $70 an hour - Full-time, Pulled from the full job description

• Health insurance
• 401(k) matching
• Dental insurance, * 401(k) matching
• Dental insurance
• Health insurance

* RAG work currently lives in Next/Nest (Node); none in Python at present. * Preference for principles over specific vendor experience; candidates with adjacent platform knowledge can adapt.