Data Scientist / Principal Investigator

CACI International•Denver, CO

7d•$108,400 - $227,500•Onsite

About The Position

CACI's is seeking a staff-level Data Scientist/Researcher to serve as a Principal Investigator and Technical Leader in our Denver, CO AI/ML Research and Integration Team. This is a hands-on role for someone who can shape ambitious research, lead teams through technical uncertainty, and turn promising prototypes into secure, scalable capabilities with measurable mission and business impact. Our IRAD team researches and develops mission-focused AI and software solutions supporting Defense, National Security, Intelligence, and other customers. We work at the intersection of applied research and product delivery—identifying high-value problems, testing new approaches, building working systems, and partnering with delivery organizations to transition successful capabilities into operational use. The Principal Investigator will help define the technical direction of a portfolio that may span generative AI and large language models, retrieval-augmented generation, natural language processing, computer vision, information extraction, recommender systems, knowledge graphs, agentic systems, and MLOps. The right candidate combines scientific judgment, strong software and systems instincts, and the ability to build alignment across researchers, engineers, customers, and business leaders.

Requirements

Bachelor's degree in computer science, data science, statistics, mathematics, engineering, or another relevant technical or quantitative discipline. An equivalent combination of education and experience may be considered.
Minimum of 7 years of professional experience.
Demonstrated experience serving as a Principal Investigator, research lead, technical lead, lead architect, or comparable role with responsibility for technical scope, team execution, risk, and outcomes.
Deep practical experience applying machine learning to real-world problems, with significant depth in two or more areas such as generative AI/LLMs, NLP, computer vision, information extraction, recommender systems, graph or semantic methods, supervised or unsupervised learning, or reinforcement learning.
Strong Python skills and experience with modern AI/ML libraries and frameworks such as PyTorch, TensorFlow, scikit-learn, Hugging Face, or comparable tools.
Minimum Clearance Required to Start: TS/SCI
Candidates must possess an active Top Secret/SCI clearance.

Nice To Haves

Candidates outside Colorado should be willing to relocate.

Responsibilities

Serve as Principal Investigator and technical lead for a portfolio of applied AI research efforts; define research questions, technical approaches, milestones, evaluation plans, risks, and transition criteria.
Identify and shape new research opportunities with technical leaders, customers, business development teams, and program organizations; contribute to white papers, proposals, technical volumes, estimates, and funding briefings.
Remain hands-on: design experiments, build prototypes, review code and models, troubleshoot difficult technical issues, and establish credible technical baselines.
Architect end-to-end AI systems that integrate models, data pipelines, APIs, search or retrieval services, orchestration, user applications, and deployment infrastructure.
Translate ambiguous mission or business needs into executable research plans and make sound build, buy, adapt, and reuse decisions under incomplete information.
Lead cross-functional teams of data scientists, software engineers, platform engineers, domain experts, and user-experience practitioners; delegate effectively while maintaining accountability for technical quality and outcomes.
Develop other technical staff through mentoring, design and code reviews, modeling guidance, constructive feedback, and clear engineering expectations.
Drive the transition of successful prototypes into secure, maintainable, production-grade capabilities by applying modern software engineering, MLOps, CI/CD, testing, documentation, and artifact-management practices.
Establish rigorous and reproducible evaluation methods, including appropriate baselines, quantitative and qualitative metrics, error analysis, operational constraints, and responsible-AI considerations.
Communicate complex technical work clearly to technical and nontechnical audiences through demonstrations, reports, roadmaps, decision briefings, and customer engagements.
Track relevant advances in commercial and open-source AI, assess their applicability and risk, and turn useful discoveries into reusable architectures, components, and practices for teams across the company.