AI Engineer – Clinical AI Platform

Medisolv•Clarksville, TN

16h

About The Position

We are looking for an AI Engineer to help build and evolve the core application platform behind our clinical AI product. This is not a typical CRUD app role, and it is not purely a data engineering role either. You will work on the application logic, orchestration, operational tooling, and system integrations that allow our platform to process clinical documents, run graph-based AI workflows, and produce reliable outputs across multiple health systems. You will work on software that has to reconcile messy real-world inputs, support multiple customers with different requirements, and make AI behavior operationally trustworthy. The work spans architecture, product behavior, cloud systems, LLM workflows, and internal tooling. It is highly practical, technically challenging, and directly connected to real clinical use cases. This role is ideal for an engineer who likes owning real product behavior end to end: application architecture, workflow orchestration, reliability, developer ergonomics, cloud integrations, and the practical realities of shipping AI-powered software in healthcare.

Requirements

You are an owner. You take initiative, move work forward, and are comfortable owning problems from discovery through implementation.
You have a strong bias for action and know how to make progress in ambiguous situations without losing sight of quality, reliability, or the needs of the end user.
Systems thinker and builder. You can break down complex technical problems, design practical solutions, and build with maintainability in mind.
You understand how application logic, workflows, infrastructure, and internal tooling work together, and you make thoughtful design choices that support scale and long-term flexibility.
Collaborative and pragmatic. You work effectively across engineering, product, and clinical teams, balancing technical rigor with practical execution.
You communicate clearly, incorporate feedback, and build strong working relationships that help move complex work forward.
Analytical problem solver. You know how to diagnose issues methodically, evaluate tradeoffs, and make sound decisions in complex systems.
Whether the challenge is workflow behavior, model output quality, observability, or production reliability, you can identify root causes and turn insights into practical improvements.
Continuous improvement mindset. You look for better ways to build, test, validate, and operate software.
You care about reducing friction for both users and internal teams, and you use automation, tooling, and process improvements to make the platform more reliable, scalable, and easier to maintain over time.
All candidates must successfully pass a pre-employment background check and be legally authorized to work in the United States.

Nice To Haves

You will onboard and get to know the people, products and departments that make Medisolv run.
Complete onboarding across engineering, product, clinical, and data teams; build working relationships with key partners and understand how the Clinical AI platform supports customer outcomes.
Get familiar with the platform architecture, core services, data flows, deployment model, and operational tooling used to run AI workflows in production.
Set up the local development environment, access required systems, and push meaningful code changes to become productive in the codebase.
Learn the clinical document processing and graph-based workflow lifecycle, including how inputs are retrieved, transformed, evaluated, and written back to the platform.
Review current reliability, testing, observability, and prompt or workflow validation practices to understand how quality is maintained across customer-specific configurations.
Own an important area of the Clinical AI platform end to end, from design and implementation through operational readiness and ongoing improvement.
Design and ship enhancements that improve multi-tenant scalability, maintainability, and configurability across customers, facilities, and clinical registries.
Improve internal tooling and operator workflows so engineering and clinical teammates can evaluate nodes, troubleshoot issues, and manage configuration changes more efficiently.
Lead root-cause analysis and reliability improvements for production issues, using observability data and test results to harden the system.
Help raise engineering quality by contributing reusable patterns, cleaner abstractions, and better developer ergonomics across the codebase.
Be a trusted technical owner for a significant portion of the Clinical AI platform, consistently delivering reliable, maintainable, and high-impact product behavior.
Drive architectural improvements that make the platform more scalable, observable, and resilient as new AI workflows, customers, and registries are added.
Demonstrate measurable impact on product quality and operational trustworthiness through stronger validation, reduced incidents, improved workflow performance, or better model-driven outputs.
Influence cross-functional roadmap decisions by bringing sound engineering judgment to product, clinical, and platform tradeoffs.
Help shape how the team builds AI-powered software in healthcare by contributing best practices for testing, evaluation, reliability, and responsible production operations.

Responsibilities

Build and maintain the runtime workflows that retrieve clinical data, execute graph-based reasoning pipelines, and write outputs back to the platform in a reliable, observable way.
Improve how the application uses LLMs for document understanding, evidence generation, and structured extraction, with attention to correctness, latency, cost, and traceability.
Extend the platform so it can support new facilities, clinical registries, and customer-specific behavior through shared abstractions and designed configuration layers rather than ad hoc branching.
Improve the tooling that supports configuration updates, node evaluation, issue-driven workflows, and other internal product operations so that engineers and clinical teammates can work faster and safer.
Maintain and improve the validation, testing, and monitoring layers so changes to prompts, graph logic, or facility-specific configuration do not silently degrade output quality.
Help shape the codebase, patterns, and abstractions so the system stays understandable as we add new registries, workflows, and product surfaces.