Data Scientist​/MLOps Engineer

Thoriva.ai, an Infinite Company, is building the ML backbone of our government services solutions, and we need engineers who can own it end-to-end. As a Data Scientist / MLOps Engineer, you will design, deploy, and sustain production-grade AI/ML solutions that detect fraud, waste, and abuse and that improve operations across government functions. If you thrive at the intersection of rigorous data science and operational excellence, and want your models to matter, this role is built for you.

Thoriva.ai is an AI technology company delivering intelligent, data-driven solutions to transform government operations. Our work sits at the intersection of AI, strategic innovation, operational execution, and trusted partnerships. We combine AI solutions, commercial expertise, and business logic to address government and federal agencies’ biggest challenges.

As a Data Scientist / MLOps Engineer for the State & Local sector, you will design, build, and operationalize AI/ML solutions across Thoriva's enterprise platform, owning the full machine learning lifecycle from data exploration and model development through production deployment, monitoring, and retraining.

Your work will focus on fraud detection, anomaly detection, eligibility validation, and risk scoring in public-sector environments, and you will collaborate closely with product, engineering, data, and compliance teams to deliver explainable, auditable, and mission-ready AI solutions that meet the rigorous standards of government oversight.

• Design, build, test, and continuously improve machine learning models for fraud detection, waste identification, abuse pattern recognition, eligibility verification, anomaly detection, risk scoring, and program‑integrity use cases across government benefit and compliance programs.
• Perform structured and semi‑structured data exploration, feature engineering, model training, validation, and rigorous performance evaluation against real‑world agency requirements.
• Develop and maintain end‑to‑end ML pipelines spanning data ingestion, feature stores, model training, deployment, monitoring, and automated retraining cycles.
• Deploy ML models as batch scoring jobs, selecting the right deployment pattern based on business needs, latency requirements, and system architecture.
• Own model versioning, experiment tracking, reproducibility, model registry management, and structured release management practices.
• Partner directly with data engineers to ensure that upstream data pipelines deliver clean, trusted, lineage‑tracked, and well‑documented data ready for model development and validation.
• Build explainable model outputs, confidence scores, risk indicators, reviewer queues, and audit‑ready evidence packages that meet public‑sector accountability and transparency standards.
• Monitor production models continuously for data drift, model drift, accuracy degradation, bias signals, latency issues, and operational reliability — and drive remediation when thresholds are breached.
• Collaborate with SMEs, product, engineering, security, and domain teams to translate complex program‑integrity requirements into AI/ML solutions that are technically sound and operationally sustainable.
• Produce high‑quality technical documentation including model cards, validation notes, deployment runbooks, and stakeholder‑facing summary materials.
• Work alongside security and compliance teams to ensure all AI/ML solutions satisfy privacy regulations, governance policies, auditability requirements, and public‑sector data handling standards including PII and PHI protections.

• 5–8 years of hands‑on experience in data science, machine learning engineering, MLOps, data engineering, or a closely related discipline.
• Demonstrated proficiency building machine learning models in Python using libraries such as Scikit‑learn, XGBoost, Tensor Flow, PyTorch, or comparable frameworks.
• Solid working knowledge of supervised and unsupervised learning techniques, including classification, regression, clustering, anomaly detection, and risk scoring methodologies.
• Proven experience deploying ML models into production or…