The Opportunity:
As a CACI AI/ML Engineer to develop, deploy, and maintain custom artificial intelligence and machine learning models. This role sits inside the Special Operations Command Europe (SOCEUR) Hybrid Threat Action Platform (HTAP) program and is responsible for the end-to-end lifecycle of ML models on the Google Cloud Platform (GCP). The individual will build robust MLOps infrastructure, integrate vector databases, and optimize model inference to transition prototypes from isolated research environments into highly available production systems. The individual will lead the development of Agentic AI and large language model (LLM) capabilities across multiple defense disciplines including object detection, natural language processing (NLP), and time-series prediction. The individual will interact regularly with stakeholders ranging from technical leads at NASIC/DOE, HTAP engineers, to mission owners supporting Allied, partner, and US operations.


Responsibilities:

• Architect, build, and deploy machine learning and Generative AI models utilizing Kubernetes, GCP's Vertex AI, and advanced infrastructure like Tensor Processing Units (TPUs).
• Design retrieval-augmented generation (RAG) frameworks, manage multi-agent workflows, and fine-tune LLMs to solve complex, multi-layered defense and operational problems.
• Transition ML prototypes from isolated Jupyter notebooks to highly available, low-latency, and secure scaled production environments.
• Design and deploy model endpoints featuring strict security checks, error handling, retries, and fallback mechanisms.
• Establish end-to-end continuous integration and continuous deployment (CI/CD) pipelines to ensure real-time inference and seamless model updates in tactical to strategic cloud environments.
• Implement evaluation frameworks and guardrails to eliminate logical errors, hallucinations, and biases in automated operational decision-making.
• Implement rigorous model governance, ensuring version control, auditability, and Explainable AI (XAI) standards are met.
• Optimize infrastructure for high-performance computing (HPC) workloads and Edge AI deployments.
• Act as a technical advisor to HTAP mission owners and stakeholders, driving AI/ML alignment, technical project strategy, and coaching teams on cloud ML integration.

Qualifications:

Required: 

·Must be a US Citizen possessing an active TS/SCI security clearance.

·Bachelor's degree in computer science, Artificial Intelligence, Data Science, or a related field.

·8 years of experience in full-stack software development (Python or C ) with a focus on designing and testing production-grade software.

·5 years of dedicated experience in ML engineering, model deployment, and ML infrastructure optimization.

·Hands-on expertise utilizing GCP data and ML services, specifically Vertex AI, BigQuery, Dataflow, and Cloud Storage.

·2 years of experience with Generative AI technologies, including LLMs, RAG, embedding models, and vector databases (e.g., Vertex AI Vector Search).

·1 years of experience building and scaling Agentic workflows/prompt engineering.

·Proficiency with core data science and ML libraries (e.g., PyTorch, TensorFlow, Keras, scikit-learn, Pandas, NumPy, OpenCV).

·Experience contributing to enterprise codebases using version control (Git, GitLab, Bitbucket) and CI/CD pipelines.

·DoD 8570.01-M / DoD 8140 IAT Level II (or higher) baseline certification (e.g., CompTIA Security , CISSP).

·Google Cloud Professional Machine Learning Engineer certification (or equivalent GCP Professional certification).

Desired:

• TensorFlow Developer Certificate.
• Familiarity with Agentic and LLM frameworks such as LangChain, HuggingFace, Model Context Protocol (MCP), or Microsoft Agent Framework.
• Experience utilizing hardware-acceleration frameworks, specifically compiling and optimizing code for Google TPUs (utilizing JAX and XLA) as well as NVIDIA GPUs (utilizing CUDA, CuPy, Numba, or the RAPIDS/CuDF ecosystem).
• Experience in application deployment and container orchestration (Docker, Podman, Kubernetes, Rancher).
• Familiarity with multi-cloud environments, including AWS resources (EC2, S3, Lambda).
• Deep understanding of NLP algorithms (e.g., BERT) and specialized ML fields such as Speech/Audio processing or Reinforcement Learning

Candidate will qualify for:

• Moving expenses to relocate household from CONUS to Stuttgart, Germany
• Dependent Education
• LQA - Living Quarters Allowance

What You Can Expect:

 A culture of integrity.

At CACI, we place character and innovation at the center of everything we do. As a valued team member, you’ll be part of a high-performing group dedicated to our customer’s missions and driven by a higher purpose - to ensure the safety of our nation.

An environment of trust.

CACI values the unique contributions that every employee brings to our company and our customers - every day. You’ll have the autonomy to take the time you need through a unique flexible time off benefit and have access to robust learning resources to make your ambitions a reality.

A focus on continuous growth.

Together, we will advance our nation's most critical missions, build on our lengthy track record of business success, and find opportunities to break new ground - in your career and in our legacy.

Pay Range:

There are a host of factors that can influence final salary including, but not limited to, geographic location, Federal Government contract labor categories and contract wage rates, relevant prior work experience, specific skills and competencies, education, and certifications. Our employees value the flexibility at CACI that allows them to balance quality work and their personal lives. We offer competitive compensation, benefits and learning and development opportunities. Our broad and competitive mix of benefits options is designed to support and protect employees and their families. At CACI, you will receive comprehensive benefits such as; healthcare, wellness, financial, retirement, family support, continuing education, and time off benefits.

The proposed salary range for this position is: