Research Engineer - AI

## Job Summary Support applied research across the Orbis AI focus areas, including agentic AI, knowledge representation, dynamic ontologies, autonomous analysis, and high-performance analytics. Work alongside senior research engineers on questions with real product stakes. You will reproduce baselines, build evaluation datasets and harnesses, run experiments, and write up what actually happened. This role exists to bring strong early-career engineers into applied AI research and grow them toward owning a research area. ## Key Responsibilities - Reproduce baselines and published methods, and document precisely where your results diverge from what was reported. - Build and maintain evaluation datasets and harnesses, including cases that capture known failure modes. - Run experiments designed with senior research engineers, keeping records complete enough that someone else can rerun them. - Perform error analysis, characterizing what the system got wrong and why, and report it without smoothing it over. - Implement well-scoped prototype components with design direction and code review from senior engineers. - Summarize relevant literature and open-source work, and explain how it bears on the question at hand. - Write up results, negative results included, and present them in team research reviews. - Travel occasionally, up to 10%, for conferences, training, and team collaboration. ## Required Qualifications - 0-2 years of professional experience - Bachelor's or Master's degree in Computer Science, Machine Learning, or a related technical field, or equivalent demonstrated ability - Solid programming fundamentals in Python, with code you can walk through and defend - Working understanding of machine learning and LLM fundamentals, including evaluation basics, prompting, embeddings, and retrieval, from coursework, projects, internships, or professional work - Ability to read a technical paper and explain what it claims, how it was evaluated, and where it might not hold - Intellectual honesty about results, including those that undercut a preferred hypothesis - Willingness to travel occasionally, up to 10% ## Desired Qualifications - Research experience from a thesis, lab, internship, or open-source project - Hands-on project experience with retrieval-augmented generation, agent frameworks, or fine-tuning - Exposure to graph databases, ontologies, or information extraction - Experience with experiment tracking and evaluation tooling such as notebooks, dataset construction, or scoring pipelines - Exposure to containerized environments (Docker) and cloud or GPU compute - Interest in OSINT, intelligence, or investigative research workflows ## Physical Requirements - Prolonged periods of sitting at a desk and working on a computer - Participation in virtual and in-person meetings - Ability to attend planned meetings and/or work in classified spaces for extended periods within the specified work regions - Travel: Occasional, up to 10%, for conferences, training, and team collaboration