Data Scientist AI / ML

Inizio Partners Corp•Washington D.C., WA

1d•Remote

About The Position

This is a long-term (12+ months) remote position based in the U.S. requiring U.S. Citizenship and the ability to obtain and maintain a Public Trust clearance. The role involves partnering with stakeholders to define and deliver AI, machine learning, and advanced analytics use cases, translating business needs into scalable data science solutions. You will design and develop machine learning models and analytical approaches for search, discovery, anomaly detection, fraud detection, and insight generation across structured and unstructured data. A key aspect of the role is using Generative AI and foundation models to enhance anomaly and fraud detection workflows. You will also build and implement natural language processing, semantic search, entity resolution, and relationship analytics capabilities. Collaboration with data engineers, cloud engineers, investigators, and business stakeholders is essential for integrating models into production environments using AWS-native services. The position requires operationalizing data science solutions within an AWS data lakehouse and developing model evaluation frameworks that ensure AI outputs are transparent and actionable. You will operate within an Agile delivery model and communicate findings clearly to both technical and non-technical audiences. Contributions to solution design, proposal support, technical documentation, and thought leadership are also expected.

Requirements

Bachelor's or Master's degree in Data Science, Computer Science, Statistics, Mathematics, Engineering, or a related field.
A minimum of 4 years of experience in data science, machine learning, artificial intelligence, or applied analytics roles.
U.S. Citizenship required and the ability to obtain and maintain a Public Trust clearance.
Experience developing and applying machine learning models in one or more of the following areas: Natural Language Processing (NLP), Semantic search or information retrieval, Entity resolution or relationship modeling, Anomaly or outlier detection, Fraud, risk, waste, abuse, or suspicious-pattern detection, Graph analytics or network analysis.
Skills and experience designing or implementing anomaly and fraud detection analyses using methods such as classification, clustering, isolation-based techniques, autoencoders, time-series analysis, behavioral analytics, link analysis, rules-based detection, or ensemble modeling.
Demonstrated competency working with AWS-native data, analytics, AI, and machine learning services, such as Amazon S3, AWS Glue, AWS Lake Formation, Amazon Athena, Amazon EMR, Amazon Redshift, Amazon SageMaker, Amazon Bedrock, AWS Lambda, AWS Step Functions, Amazon OpenSearch Service, Amazon ECS or Amazon EKS, and Amazon CloudWatch.
Demonstrated competency working with AWS data lakehouse architectures, including Amazon S3-based data lakes, Apache Iceberg or similar open table formats, centralized metadata catalogs, governed data access, schema evolution, partitioning, data quality, and scalable query and transformation patterns.
Demonstrated competency using Generative AI, large language models, or foundation models for anomaly and fraud detection analysis, including alert enrichment, investigative summarization, contextual reasoning, evidence synthesis, pattern explanation, and analyst assistance.
Experience working with large-scale structured and unstructured data, particularly document-based datasets such as text, PDFs, images, extracted fields, and metadata.
Experience using metadata, engineered features, embeddings, event or transaction histories, and relationship data to support analytics and modeling.
Strong proficiency in Python for data science and machine learning, including libraries such as Pandas, NumPy, Scikit-learn, PyTorch, or TensorFlow, along with strong SQL skills.
Experience building scalable data processing and machine learning pipelines using AWS-native services and open-source frameworks such as Apache Spark.
Experience integrating models into production environments through APIs, batch pipelines, event-driven workflows, streaming pipelines, or embedded analytics applications.
Understanding of model evaluation, validation, monitoring, drift detection, bias assessment, explainability, and performance measurement.
Understanding of operational challenges in anomaly and fraud detection, including class imbalance, changing patterns, false positives, data leakage, concept drift, and feedback-loop design.
Ability to apply responsible AI, privacy, security, and governance practices when using Generative AI with sensitive or regulated data.
Strong communication skills and the ability to translate analytical outputs, detected risks, and model results into actionable insights.
Experience working in cross-functional, matrixed teams in an Agile environment.

Nice To Haves

Experience developing production-grade anomaly detection, fraud detection, waste and abuse analytics, risk-scoring, or investigative decision-support solutions.
Experience using Amazon Bedrock foundation models or comparable enterprise Generative AI capabilities for analytics, anomaly detection, fraud investigation, or case-management workflows.
Experience implementing Retrieval-Augmented Generation (RAG), prompt engineering, agentic workflows, guardrails, foundation-model evaluation, and responsible AI controls for enterprise use cases.
Experience building AI-enabled search solutions, including semantic search, hybrid search, document retrieval, embeddings, and ranking models using AWS-native services.
Experience with multimodal data processing involving text, images, documents, metadata, and other data types.
Familiarity with OCR/ICR and document intelligence pipelines using services such as Amazon Textract and Amazon Bedrock.
Experience using Amazon SageMaker for feature engineering, model development, training, deployment, monitoring, and machine learning lifecycle management.
Experience using Amazon OpenSearch Service, graph databases, or graph-processing frameworks for relationship mapping, link analysis, and suspicious-network detection.
Experience developing explainable AI solutions, including confidence scoring, reason codes, feature attribution, evidence traceability, and analyst-friendly explanations.
Experience designing analytics dashboards, alert-management interfaces, investigative reporting, or case-triage solutions for end users.
Knowledge of AWS security and governance practices, including IAM, encryption, logging, data classification, least-privilege access, and handling sensitive or regulated information.
Previous experience supporting federal clients or working in regulated environments.
Consulting experience or experience in a client-facing delivery role strongly preferred.
Experience supporting training, user enablement, or adoption of analytics capabilities across teams.
Familiarity with graph-based analytics, ontology-driven models, knowledge graphs, or relationship mapping.

Responsibilities

Partner with stakeholders to define and deliver AI, machine learning, and advanced analytics use cases, translating business needs into scalable data science solutions.
Design and develop machine learning models and analytical approaches to support search, discovery, anomaly detection, fraud detection, and insight generation across structured and unstructured data.
Develop anomaly and fraud detection analyses using supervised, unsupervised, semi-supervised, statistical, and graph-based techniques, including outlier detection, behavioral profiling, risk scoring, pattern detection, and relationship analysis.
Use Generative AI and foundation models to enhance anomaly and fraud detection workflows through alert enrichment, case summarization, contextual analysis, evidence synthesis, pattern explanation, investigative hypothesis generation, and analyst decision support.
Build and implement natural language processing, semantic search, entity resolution, and relationship analytics capabilities to enable advanced information retrieval and identification of suspicious patterns and connections.
Leverage document-based data, including OCR/ICR outputs, metadata, images, extracted fields, and free text, to support downstream analytics, search, anomaly detection, and fraud analysis.
Collaborate with data engineers, cloud engineers, investigators, and business stakeholders to integrate models and analytical capabilities into production environments using AWS-native services.
Develop and operationalize data science solutions within an AWS data lakehouse, including scalable data preparation, feature engineering, model training, inference, monitoring, and analytics.
Develop model evaluation frameworks, confidence and risk scoring, explainability, traceability, and human-in-the-loop review approaches to ensure AI outputs are transparent, actionable, and suitable for investigative and operational use.
Evaluate anomaly and fraud detection models using appropriate measures, including precision, recall, F1 score, false-positive rate, detection rate, ranking quality, and business impact.
Support the development of dashboards, reporting, alerting, and investigative workflows that drive operational insights and informed decision-making.
Operate within an Agile delivery model, contributing to sprint planning, experimentation, model iteration, and incremental solution delivery.
Communicate findings, risk indicators, model limitations, and recommendations clearly to technical and non-technical audiences, including client stakeholders.
Contribute to solution design, proposal support, technical documentation, and thought leadership in AI, analytics, anomaly detection, and fraud detection.