Principal AI Platform Engineer
Dallas, TX (Hybrid)
Full time |
Role Overview
We are seeking a Principal AI Platform Engineer to build our AI software factory. This is a hands-on individual contributor role. You will write code every day, ship the first working version of our agent platform yourself, and then teach engineering teams to build on it. You will have no direct reports at the start. The platform will run agents against our production codebase, with people writing specifications and validating results and agents doing most of the implementation. You will also serve as the senior technical authority on AI built into our products.
Our primary stack is .NET and Azure, with Azure DevOps, Kubernetes, Azure AI Foundry and MCP-based tooling, alongside other languages and frameworks. The agent platform must work across all of them. This is a greenfield effort, so you will set the standards rather than inherit them.
Is This Role for You?
This role is for someone who builds. You'll set standards, shape how teams use AI and influence the roadmap, but you'll do it through working code and reference implementations, not decks, review boards or org charts. If your recent work has mostly been leading teams, advising executives or designing solutions that others implement, this probably isn't the right match. Expect to spend most of your time coding, especially in your first 6 months.
Key Responsibilities
AI Platform Architecture and Development
- Design, write and operate the agent platform that runs autonomously against our production codebase, taking work from specification to pull request with human review.
- Implement agentic patterns, orchestration and tool integration for real-world engineering workflows.
- Make the platform effective across our full technology stack, not only one language or framework.
- Define the agent templates and shared framework that every team uses.
- Own the AI lifecycle from concept through deployment, operation and iteration on Azure and Azure AI Foundry.
- Shape how work is handed to agents and review the results, from concept to specification through pull request.
Developer Experience and Integration
- Partner with application teams to integrate AI capabilities into existing products and workflows.
- Build the APIs, SDKs, MCP servers and extension points that engineering teams build on.
- Establish how MCP servers, tools and agent skills are packaged, versioned, discovered and distributed to teams, such as through an internal registry or catalog with access controls.
- Design tool access with per-tenant data scoping, so that an agent acting for one agency can reach only that agency's data.
Context and Knowledge Engineering
- Design the context, skills and retrieval patterns that make AI agents effective in large codebases, including an application-level knowledge map that tells an agent which module a ticket belongs to and what it affects.
- Decide when retrieval is the right tool and when curated context is better.
- Establish versioning, review and governance for shared AI assets.
Evaluation, Safety and Operational Quality
- Build evaluation frameworks that measure AI output quality and detect regressions.
- Define guardrails, failure handling and human-in-the-loop controls for regulated environments, including keeping PHI out of agent context, prompts and logs.
- Monitor and optimize AI systems for reliability, accuracy and cost efficiency, with token budgets, rate limits, retry caps and step limits.
- Establish observability best practices across supervised and autonomous AI usage.
Enablement and Technical Leadership
- Mentor engineers across the organization and raise AI engineering capability at scale, by building working examples in their own codebases.
- Define engineering standards and shape the AI technical roadmap with the CTO.
Innovation and Applied Research
- Evaluate emerging capabilities in LLMs, agentic AI, RAG and MCP, and bring what proves out into production.
- Advise on build, buy and partner decisions across the AI toolchain.
- Maintain a working view of commercial and open-weight models, and match model tiers to tasks by cost, latency, quality and data-handling terms.
Qualifications (Required)
- Bachelor's or advanced degree in Computer Science, Engineering or a related field, or equivalent practical experience.
- 10+ years in software engineering, with recent hands-on coding. You write code today, not only review it.
- 3+ years building and operating production AI systems in which agents call tools or change code. RAG or machine learning products alone do not meet this bar.
- Proficiency in Python and modern AI tooling (for example LangChain, LangGraph, MCP, OpenAI and Anthropic SDKs).
- Hands-on experience building MCP servers or equivalent tool interfaces, including authentication and per-tenant access control.
- Ability to make agents effective across multiple languages and technology stacks, not only one.
- Hands-on experience deploying and operating AI workloads in cloud environments (Azure preferred).
- Solid grasp of LLMs, generative AI, agentic patterns and context engineering.
- Proven RAG experience, and judgment on when retrieval beats curated context.
- Experience with AI evaluation and observability, including quality measurement, cost and token management.
- Working knowledge of handling regulated data such as PHI in AI systems.
- Demonstrated record of mentoring engineers and raising technical capability across teams.
- Evidence of your own work: a repository, code sample or demo of agent or MCP work that you personally built.
Preferred (Nice to Have)
- Experience integrating AI into C# and .NET applications.
- Experience using AI coding tools such as Cursor or Claude Code on large codebases.
- Familiarity with vector databases and retrieval tooling (for example pgvector, Pinecone, Azure AI Search).
- Exposure to LLMOps and evaluation tooling (for example Azure AI Foundry, MLflow, LangSmith, CI/CD).
- Familiarity with agent-to-agent communication standards such as A2A, and judgment on when a multi-agent design is warranted over a single agent with tools
- Experience with large legacy codebases and incremental modernization.
- Interest in healthcare, EHR or revenue cycle applications.
