Lead Agent Engineer

LTS is seeking a Lead Agent Engineer to join a small, senior engineering team applying frontier AI to one of the most consequential legacy systems still running in production today.

The mission: build agents that read, translate, and modernize a decades-old codebase that millions of people quietly depend on. The work has executive backing, real users, and a customer who knows exactly what they’re buying. Specifics shared once we’re talking.

The team is small by design. Every seat carries unusual leverage, and we hire people who are already deep in this work. We use AI tooling natively — agents in parallel, model as collaborator, no exceptions.

What You'll Do:

The Lead Agent Engineer sets the technical ceiling on what the platform can do: translation accuracy, dependency-map correctness, hallucination control, multi-agent orchestration. Capability bets at the frontier of code understanding live here. So does the prompt-and-eval iteration cycle, in partnership with the RAG & Eval Engineer.

• Own the agent layer of the platform — architecture, prompts, tool surfaces, multi-agent orchestration, and the eval loop.
• Set the technical ceiling for the platform’s core capabilities: translation accuracy, dependency-map correctness, hallucination control.
• Drive translation and dependency-mapping accuracy across an unfamiliar legacy paradigm — call graphs, data references, semantic equivalence checks.
• Take on capability bets that don’t show up on most roadmaps yet — agent work at the frontier of code understanding.
• Drive the prompt-and-eval iteration cycle in partnership with the RAG & Eval Engineer.
• Write production agent code daily; subagents and multi-agent workflows are your normal way of working.
• Establish foundations the rest of the team builds against — agent architecture, eval loop, prompt-and-tool patterns, multi-agent topology.

What We’re Looking For:

• Bachelor’s degree in Computer Science, Engineering, or a related field, plus 5 years of professional software engineering experience; equivalent experience may substitute for the degree requirement.
• Has shipped at least one production agent (real users, real failure modes) and can describe in detail what broke and how it was fixed.
• Ability to work in a fast-paced, collaborative environment.
• Working fluency across the modern agent stack: tool use, MCP, evals, RAG, streaming, multi-agent coordination.
• Strong TypeScript or Python.
• Subagents and multi-agent workflows are a normal way of working.
• Comfort with measurement, regression detection, and prompt-and-eval iteration.
• Ability to operate without a roadmap — defines the next problem worth solving and goes.
• Heavy native use of AI tooling: agents in parallel, model as collaborator.
• Demonstrated experience in a remote work environment.
• Prior architectural ownership of an agent platform; track record of establishing eval and prompt patterns for a team.

Nice to Have:

• Tooling around Claude or comparable frontier models at scale.
• Code-understanding agent work — translation, refactoring, test generation, dependency analysis.
• Background in compilers, program analysis, or static/dynamic code understanding.
• Open-source contributions to agent frameworks, MCP servers, or eval harnesses.
• Legacy-language or legacy-system exposure (rare, and a plus rather than a requirement).
• Public technical writing or conference talks on production agent work.
• Healthcare IT or legacy modernization domain experience.

What’s in it for you?

• The opportunity to support high visibility federal missions in IT and healthcare
• A culture that values innovation, growth, collaboration, and quality
• Access to cutting-edge tools and technologies
• Comprehensive benefits for you and your family
• A career path that rewards ambition and performance

If you’re ready to push boundaries, sharpen your skills, and join a team that is passionate about building what’s next, we’d love to meet you. Apply today and let’s build a future together!