AI/ML Fine-Tuning and Training Engineer

Role Description

We are seeking a hands-on and technically rigorous AI/ML Fine-Tuning and Training Engineer to design, train, fine-tune, and evaluate machine learning and large language models that power Verterim's AI agents and GRC AI products, under the direct direction of Verterim's CTO.

This role sits at the intersection of applied ML engineering, model training infrastructure, and domain-specific fine-tuning, translating the CTO's technical direction into trained, evaluated, and production-ready models that meet the accuracy, explainability, and auditability standards required for regulated GRC environments. The Engineer will work in close daily partnership with the CTO and the AI Solutions Engineering team to build and continuously improve the models underlying Verterim's multi-agent systems.

Qualifications

• 4+ years of experience in machine learning engineering, applied NLP, or LLM training/fine-tuning roles.
• Demonstrated hands-on experience fine-tuning large language models (e.g., LoRA/QLoRA, full fine-tuning, instruction tuning, or RLHF/DPO).
• Strong proficiency in Python and modern ML frameworks (PyTorch, Hugging Face Transformers/PEFT/TRL, or equivalent).
• Solid understanding of model evaluation methodology, including benchmark design, offline/online evaluation, and regression testing.
• Proven ability to execute under strong technical leadership while translating strategy into working, production-ready models.
• Strong communication skills with the ability to operate effectively with executive-level technical leadership.

Requirements

• Fine-tune and train large language models and other ML models as directed by Verterim's CTO, translating technical strategy into trained, deployable model artifacts.
• Design and execute fine-tuning pipelines (full fine-tuning, LoRA/QLoRA, instruction tuning, RLHF/DPO) tailored to GRC, risk, compliance, and audit use cases.
• Curate, clean, and structure training and evaluation datasets, including synthetic data generation and domain-specific labeling workflows.
• Optimize model performance across accuracy, latency, cost, and resource utilization for production deployment.
• Build and maintain rigorous evaluation harnesses, benchmarks, and regression suites to validate model quality prior to release.
• Ensure fine-tuned models meet enterprise expectations for accuracy, explainability, traceability, and defensibility in regulated environments.
• Detect and mitigate bias, hallucination, and drift through structured testing and continuous monitoring.
• Document model lineage, training data provenance, hyperparameters, and evaluation results to support audit and governance requirements.
• Manage end-to-end training infrastructure and workflows, including compute provisioning, experiment tracking, and versioning of models and datasets.
• Partner with AI Solutions Engineering to integrate trained models into RAG pipelines, agent orchestration frameworks, and knowledge-base systems.
• Establish reproducible, automated pipelines for retraining, fine-tuning refreshes, and model promotion from experimentation to production.
• Maintain technical documentation, training runbooks, and internal enablement artifacts for models in production.
• Serve as the primary technical point of contact for model training and fine-tuning matters between the CTO, engineering, and product teams.
• Ensure model development aligns with Verterim's internal AI governance model, quality funnel standards, and responsible AI principles.
• Partner with the CTO and Product Management to translate business, regulatory, and customer requirements into model training objectives.
• Support customer-facing technical discussions where deep model architecture or performance detail is required.

Benefits

• Fully Remote
• Direct report to the CTO
• Minimal travel (technical reviews, select industry/conference events)