TixelJobs
N
nanvia muse

Software Engineer, AI Solutions

REMOTE$121K - $133K/yrPosted 3d ago
ML EngineerMid LevelFull-time#remote

About the Role

POSITION: Software Engineer, Applied AI





DIRECT REPORT: Sr. Director, AI and Architecture





COMPENSATION: Pursuant to state regulations, the salary range is $120,800\-$133,300 plus bonus potential. Eligible to participate in the company benefits plan (https://cardinalgroup.com/wp\-content/uploads/2025\-Benefits\-Guid\-Eng\_SPA.pdf). We offer health, vision, dental, and pet insurance. We offer a 401(k) retirement plan, student loan assistance, licensing and continuing education reimbursement, parental leave, and housing allowances or gifts.





SUMMARY



The Software Engineer, Applied AI is responsible for designing, building, and supporting production software and AI solutions that solve business problems across Cardinal. Working across the full software development lifecycle, this role owns solutions end\-to\-end, including web applications, APIs, integrations, and AI agents. The engineer works AI\-natively, using agentic development tools such as Claude Code and GitHub Copilot as everyday collaborators, while bringing the engineering mindset and computer science fundamentals needed to direct, review, and verify AI\-generated work so that what ships is secure, tested, and maintainable. Operating with a high degree of independence, this role moves quickly from prototype to production, partners with senior engineers on architecture, and helps raise the technical bar across the team.





RESPONSIBILITIES (Including but not limited to)


* Translate business requirements into well\-designed technical solutions, including web applications, APIs, integrations, automated workflows, and AI agents, and own their delivery end\-to\-end, from rapid prototypes that validate ideas with stakeholders to hardened production releases.
* Build backend services and RESTful APIs using Python and FastAPI, and front\-end applications using TypeScript and Angular.
* Use agentic development tools (e.g., Claude Code, GitHub Copilot) as a core part of the daily workflow, from planning and specification through implementation, testing, and documentation, including delegating well\-scoped tasks to coding agents and rigorously reviewing the results.
* Build and maintain the team's shared AI development assets, such as agent instruction files, reusable skills and commands, and Model Context Protocol (MCP) servers, to make AI\-assisted development faster and more consistent.
* Design, build, and operate AI agents and LLM\-powered features using frameworks such as Google Agent Development Kit (ADK) or LangGraph, applying tool calling, retrieval\-augmented generation (RAG), structured outputs, and multi\-agent patterns where appropriate.
* Bring the same rigor to AI features that the team applies to conventional software: build evaluation datasets and automated evals, instrument tracing and monitoring, and track quality, cost, and latency in production.
* Write and maintain automated tests (unit, integration, end\-to\-end) and contribute to CI/CD pipelines in GitHub Actions, including automated quality gates and AI\-assisted code review.
* Deploy and operate solutions on Google Cloud Platform (GCP), including Cloud Run, Cloud SQL, BigQuery, and Vertex AI, using containers and infrastructure\-as\-code where appropriate, and participate in production support, including troubleshooting, root\-cause analysis, and reliability improvements.
* Design data models and build integrations with internal and third\-party systems using APIs, webhooks, and event\-driven messaging, ensuring reliable data flows.
* Apply secure\-by\-design practices, including authentication and authorization, secrets management, least\-privilege access for agents and their tools, protection of resident and team member data, and mitigation of AI\-specific risks such as prompt injection.
* Contribute to architecture and technical design discussions alongside senior engineers, documenting decisions and trade\-offs.
* Perform thorough code reviews of both human\- and AI\-authore
Share