Comand AI's mission is to build next-generation C2 software with real users in real deployments in the field.
We are fully indexed on the success of every single user, which means the software you ship runs inside live defense operations, not a benchmark.
You'd be joining the engineering team working directly with the leadership.
You take a problem, explore it, come back with proposals, and iterate; nobody hands you a fully specified ticket.
Most of the team genuinely values that exposure.
It isn't a dealbreaker if you're very strong but less drawn to it, but expect real user contact to be part of the job.
What this looks like in practice:Own end-to-end features on Prevail, our operational planning platform, from scoping through shippingWork on real geospatial problems: everything in the product lives on a map, so you'll move data between backend and frontend through tile servers and lean on PostGIS for geo queries that need to be genuinely smartBuild and harden ML/LLM workflow orchestration: most product workflows carry an LLM or ML step, so you'll design for retries, queueing, and long-running jobs with LangGraph and LangChainDesign for on-prem and appliance deployment: our default target isn't cloud-native, it's hardware plus software shipped as an appliance, so you're thinking.
We want people who reason from first principles rather than lean on memorized patterns, and who ship rather than over-plan.
Must-have:Full-stack experience preferred; if not full-stack, a genuinely strong backend lean is requiredComfortable owning a feature end to end: scoping it, building it, and deciding what not to build, without a PM pre-slicing the workSolid Python (we use FastAPI) and TypeScriptEU citizenship (contractual requirement tied to our defense contracts)Rea.