Senior Agentic AI Software Engineer – Python, Go & Vertex AI

Keyhole Software is hiring Senior Agentic AI Software Engineers with strong software engineering fundamentals and hands-on experience building agentic systems to support a long-term enterprise healthcare client engagement. This consulting role is ideal for experienced engineers who understand how to move beyond basic LLM integrations and build reliable, maintainable agent workflows and supporting infrastructure.

• Title: Senior Agentic AI Software Engineer
• Location: Remote within the United States. Preference will be given to candidates located in the Kansas City metro area, followed by candidates residing elsewhere in Kansas or in Nebraska, Texas, Florida, or Arizona.
• Eligibility: Candidates must be authorized to work and reside in the United States. No Corp-to-Corp (C2C) or Contract-to-Hire (C2H). At this time, Keyhole Software is unable to provide visa sponsorship.

Keyhole Software is seeking experienced Senior Agentic AI Software Engineers to join our team. Our consultants are full-time members of Keyhole Software, working across a variety of long-term client engagements throughout their careers.

For this engagement, you’ll help build Data Assist, a purpose-built agentic harness designed to assist data engineers with tasks such as data extraction and code generation. The solution will integrate with the client’s existing model gateway and agent infrastructure on Google Cloud Platform.

The technology stack is evolving and generally includes Python, Go, Google Cloud Platform, and Vertex AI, along with specialized tools selected as the architecture develops. We’re looking for adaptable engineers who can work effectively in an emerging technical space, evaluate options thoughtfully, and build reliable systems without depending on a single framework or fixed implementation pattern.

Strong fundamentals and meaningful agentic engineering experience are essential. You should understand the engineering challenges involved in agent orchestration, tool use, context management, structured outputs, evaluation, observability, reliability, and safeguards. Previous experience building an agent harness or working on a data or data-platform engineering team would be especially valuable.

If you enjoy building practical AI systems, working through evolving technical challenges, and creating tools that make other engineers more effective, we’d love to talk with you.

• Design, build, test, and refine components of a purpose-built agentic harness for data engineering workflows.
• Develop production-quality services and agent workflows using Python, Go, and related technologies.
• Integrate the harness with an existing model gateway and agent infrastructure on Google Cloud Platform.
• Build and maintain agent capabilities involving tool use, structured outputs, task orchestration, context management, and code generation.
• Develop integrations that allow agents to safely interact with data systems, engineering tools, APIs, and internal services.
• Improve the reliability, observability, testability, and performance of agentic workflows.
• Design evaluations and testing strategies for non-deterministic AI behavior, failure modes, and generated outputs.
• Troubleshoot issues across application code, model interactions, tools, prompts, cloud services, and supporting infrastructure.
• Apply appropriate safeguards, access controls, and engineering judgment when agents interact with sensitive systems or data.
• Collaborate with client engineers, data engineers, and technical stakeholders to understand workflows and translate them into useful agent capabilities.
• Evaluate emerging libraries, frameworks, and implementation approaches as the technology stack evolves.
• Write clean, maintainable, well‑tested code and…