At Jaxon, we’re focused on making AI trustworthy for mission-critical environments. Our core technology is built to enable the safe deployment of AI in high-stakes settings, particularly across the Department of Defense. DSAIL converts human policy, procedures, and doctrine into executable guardrails that constrain and verify LLM outputs.
Requirements
- Python
- Agentic orchestration libraries such as Langraph, AgentKit, SonnetSDK, etc. Custom tools, RAG, and other agentic helpers are not new shiny objects for you; they’re basic building blocks.
- Docker, Docker Compose, K8s, and related container orchestration technologies.
- Hands-on experience running local LLMs using technologies such as Ollama and Transformers.
- Have a comfort level building and training and/or fine-tuning models with techs like Torch or Tensorflow, not just calling cloud APIs (but you’re resourceful enough to recognize when you can get away with using cloud APIs).
- Claude Code, Cursor, Codex and other AI coding helpers - you know how to use them in anger, and not vibe your way into spaghetti.
- Strong software engineering skills and collaborative software development.
Responsibilities
- Design and Implement Agentic Systems: Develop multi-agent reasoning workflows using frameworks like LangGraph or custom orchestration layers. Integrate DSAIL evaluators into live inference pipelines.
- Bridge Symbolic and Neural Logic: Implement AI components that connect LLM outputs to symbolic verifiers (e.g., SMT solvers) for auditable reasoning.
- Engineer End-to-End Systems: Build across the full stack — backend services, orchestration layers, internal tooling, and user interfaces — to deliver complete, production-ready AI applications, not just model endpoints.
- Develop Real-World Use Cases: Prototype, demo, and iterate on customer pilots that showcase DSAIL-powered verification in defense and enterprise settings. Collaborate with stakeholders to transform proofs-of-concept into deployable solutions.
- Build Reliable, Composable Infrastructure: Deploy containerized AI services (Docker, Compose, K8s) with strict observability and reproducibility.
- Improve Performance and Traceability: Profile latency, caching, retrieval accuracy, and reasoning reliability across hybrid neural-symbolic stacks.
Other
- Full-time
- Direct Hire. No third-party assistance is approved for this role.
- Previous experience with consulting on Defense projects involving software development, data science support, especially at service labs or joint commands.
- Full-ML-stack experience: delivering ML services from conception to production system to maintenance and iterative improvement.
- We are looking for somebody who thrives in an environment where day-to-day priorities and tasks may rapidly change. The role is not a fit if you value "business as usual".
- We value team members who are bold enough to pitch creative ideas, even if they're still rough around the edges or might not make the final cut.
- Applicants should have a bachelor's degree in Computer Science or Software Engineering with a minimum 2 years of experience in applicable roles.
