Applied AI Engineer

Plymouth, MN (On-Site)
40 hrs/week | Start: Immediate


The Opportunity

The SaaSpocalypse is real. Salesforce, Workday, ServiceNow, and Epic are all feeling it. The per-seat, legacy-workflow model is being dismantled in real time and the companies that built their moats around it are scrambling. We are not.


Think Harvey for legal, Sierra for customer service, Ramp for finance - vertical AI companies built from the ground up to displace the incumbents, not compete with them on features. We are that company for healthcare.


EWINGS, INC. is a 24-year-old firm with active Fortune 500 clients, battle-tested delivery workflows, and deep regulatory domain expertise that most software companies will never acquire. We are now converting that institutional knowledge into an AI-native operational platform purpose-built for enterprise health plans - one of the most complex, high-stakes segments in the entire enterprise software market ripe for disruption.


This is not a seat-license play. This is not a dashboard bolted onto a legacy workflow. This is vertical AI infrastructure, built AI-first from the ground up, with human-in-the-loop design at its core and agentic workflows surfacing as clear, actionable intelligence for enterprise operators. The incumbents in this space are slow, expensive, point solutions incapable of moving at AI speed. We are not.


We are well-capitalized, operationally stable, and already generating revenue. Live client pilots are underway. The platform is real, the clients are real, and the urgency is real.


You Don't Have to Move to San Francisco for This

For two decades, the most compelling platform opportunities - the ones where you build something that matters, with people who know what they are doing, with real clients and real stakes - have been concentrated on the coasts. The AI wave has made that worse, not better.


This one is in Plymouth, MN.


If you have been watching the AI wave build from the Midwest and waiting for something real to land close to home - a vertical AI platform with Fortune 500 clients, live pilots, domain depth that startups cannot fake, and a founding team that has been executing at the highest level for over two decades - this is it.


The engineers who get in early at companies like this do not look back.


Who You'll Be Working With

Most AI startups are being built by first-time founders who are learning management, sales, domain expertise, and product strategy in real time - at your expense. The pressure is real, the learning curve is steep, and the failure rate reflects it.


This is not that.


The founder and CEO leading this platform brings 24 years of executive operational, sales, and market development experience inside Fortune 500 healthcare - including UnitedHealth Group, Optum, HealthNet, Ascension, and the VA Community Care Network. This is not someone learning healthcare. This is someone who built it.


The broader team matches the ambition. Our operational and delivery leadership includes people who have sat in the same seats as our clients - direct experience inside the nation's largest health plans. That team is complemented by engineers and builders with backgrounds at industry-defining companies like Google and Amazon. You will be working with people who have built at scale before and know what that requires.


The Culture

A lot of companies say they are "AI-forward." Most of them mean they added a ChatGPT integration and declared victory. The engineers who join those teams spend the next two years watching leadership catch up to where the market already was.


We are not debating anything.


At EWINGS, learning and leveraging AI every single day is the starting point, not an aspiration. Every person on this team is expected to use AI tools actively and fluently - not because we asked them to, but because that is the only way to move at the speed this market requires.


The window of pre-adoption in this market is open right now. Our competitors are still holding meetings about whether to adopt. We are already building. That gap closes fast and the engineers who are inside it when it closes are the ones who define what the platform becomes.


What you will find here is a fast-paced, highly collaborative environment with genuine enthusiasm for what we are building, clear ownership of meaningful work, and leadership that has been doing this long enough to know what good looks like.


If that is the kind of environment that energizes you, you will fit in immediately.


About the Role

You'll design how agents actually do the work: reasoning over the data that runs an entire industry, automating workflows that used to take weeks, and building the agent infrastructure the platform runs on. If you've been bolting AI onto outdated, legacy products and want to build AI-native from the foundation up, this is that job.

What you'll do:
- Design and build agent workflows that automate operational processes across the platform.
- Build the infrastructure that lets agents reason over platform data and act on it reliably.
- Develop retrieval, context, and evaluation pipelines that keep agent output accurate and trustworthy.
- Help shape how rule-governed steps run deterministically while judgment-based steps reason freely. This is a target pattern we're building toward, and you'll have real input into it.
- Implement logging and traceability for AI inferences that affect a user-visible output or decision, so nothing the system does is unexplainable after the fact.
- Build evaluation harnesses and feedback loops to measure and improve agent performance over time.
- Integrate model and service APIs, and design the boundaries between automated and human-reviewed steps.
- Partner with full-stack and design to ship agent capabilities into the product, not into a lab.
- Use AI tooling fluently across your own workflow. This is how we work.

What you need:
- Active, daily use of AI-assisted development tools. Non-negotiable.
- Strong experience building production applications with LLMs: agent patterns, tool use, retrieval-augmented generation, structured output.
- Proficiency in Python and/or TypeScript.
- Solid software engineering fundamentals. You ship maintainable, tested code, not just notebooks.
- Experience designing prompts, context strategies, and evaluation methods for real workloads.
- Comfort integrating model and service APIs (REST, OAuth, webhooks).

Nice to have:
- Experience with agent orchestration, multi-step tool-using systems, or agent-to-agent patterns.
- Experience building evaluation, observability, or guardrail systems for AI in production.
- Familiarity with healthcare data, or regulated and compliance-sensitive environments.
- A track record of moving AI features from prototype to reliable production service.

Tech environment:
The same core platform stack as the rest of the team, with AI capabilities built into the product rather than run as side experiments. Strong emphasis on traceability and auditability of AI behavior. Specific tooling and model details are shared during the interview process.