Applied AI/ML Engineer

About the role

Oddball is seeking an Applied AI/ML Engineer to own the full lifecycle of practical artificial intelligence features within federal-facing software products. You will be responsible for designing and implementing machine learning solutions that directly improve daily operations for millions of users in the public sector. This role requires you to balance cutting-edge GenAI techniques with the constraints of production environments and federal requirements. You will translate ambiguous real-world problems into well-defined ML tasks that drive measurable impact. Collaboration with engineers, designers, and product teams will be central to your success in delivering reliable AI capabilities. You will own the decisions that bridge the gap between research prototypes and scalable, maintainable software systems.

Key facts

What you'll do

• Design and deploy machine learning and AI-powered features into production systems with high reliability and performance standards.
• Apply supervised, unsupervised, and deep learning techniques to analyze structured and unstructured data in support of federal use cases.
• Build and evaluate models for tasks such as classification, ranking, prediction, natural language processing, and anomaly detection to address specific user needs.
• Develop and integrate GenAI solutions, including large language model workflows, retrieval-augmented generation patterns, and agent-based systems.
• Translate business and user requirements into clear ML problem statements, evaluation metrics, and controlled experiments that validate impact.
• Implement robust data pipelines and feature engineering workflows to ensure high-quality model training and efficient inference at scale.
• Evaluate model performance, fairness, drift, and reliability over time, and iterate on designs based on empirical results and stakeholder feedback.
• Collaborate closely with software engineers to integrate models into APIs, services, and user-facing applications while maintaining software quality standards.
• Contribute to architecture decisions around model serving infrastructure, scalability, security, and cost optimization for federal deployments.
• Document methodologies, assumptions, and tradeoffs associated with modeling choices to support long-term maintainability and knowledge transfer.

Requirements

• Must be authorized to work in the United States and capable of obtaining necessary employment eligibility documentation.
• Candidates must be located within the DMV area, including Washington DC, Maryland, and Virginia, to enable regular in-office collaboration.
• Possess a strong foundational understanding of machine learning concepts, including model selection, training methodologies, validation strategies, and performance evaluation.
• Demonstrate prior experience building, deploying, and maintaining ML models within real-world applications that serve actual users.
• Show proficiency in Python and experience working with common ML libraries such as PyTorch, TensorFlow, and scikit-learn for model development.
• Bring experience with large language models, embedding techniques, and prompt-driven systems to solve complex task objectives.
• Familiarity with data processing tools and workflows, including but not limited to Pandas, SQL, and Spark, for managing large and diverse datasets.
• Adhere to software engineering best practices, including version control, automated testing, and participation in code reviews to ensure high-quality deliverables.
• Ability to reason about tradeoffs between model accuracy, system latency, operational cost, and long-term maintainability in production contexts.
• Strong written and verbal communication skills, with the ability to work effectively in cross-functional teams across engineering, design, and product roles.
• Willingness to perform other related duties as assigned to support team objectives and project milestones.

Nice to have

• Prior experience working within innovation, R&D, laboratories, or exploratory engineering teams focused on emerging technologies.
• Demonstrated experience deploying machine learning models to cloud platforms and managing inference workflows at scale.
• Familiarity with MLOps practices, including model monitoring, CI/CD pipelines for machine learning, and experiment tracking frameworks.
• Active participation in architectural discussions and contributions to technical strategy that influence long-term product direction.

Practical notes

This role is hybrid or remote with a requirement for in-office collaboration in the Washington DC metropolitan area. Only candidates authorized to work in the United States and located in the DMV region will be considered. No specific hours or visa sponsorship details are outlined in the source material.