> Markdown version of [/jobs/ext/3666181-security-ai-ml-engineer-lead](https://www.wearedevelopers.com/jobs/ext/3666181-security-ai-ml-engineer-lead). Every page supports `.md` or `Accept: text/markdown`. Links point to the HTML versions so they work for humans too. Agent guide: [/agents.md](https://www.wearedevelopers.com/agents.md).

---

# Security AI / ML Engineer, Lead

- **Company:** Toyota Financial Services
- **Location:** Plano, United States
- **Experience:** Expert
- **Contract:** Permanent contract
- **Skills:** Training Data, A/B Testing, Artificial Intelligence, Amazon Web Services, Continuous Integration, Intrusion Detection and Prevention, Python (Programming Language), Team Foundation Server, Performance Tuning, Azure Machine Learning, Security Information and Event Management, Software Deployment, Management of Software Versions, QLoRA, Pytorch, Evaluation Pipelines, Retrieval-Augmented Generation, Transfer Learning, Multi-Agent Systems, Prompt Engineering, Model Validation, Caching, Discretization, Data Lakes, Attention Mechanisms, HuggingFace, Transformer Architectures, Build Process, Triton Inference Server, Machine Learning Operations, Model Registry, Web Api
- **Published:** October 10, 2026
- **Apply:** https://www.dice.com/job-detail/fc3b5fe5-82ca-468c-bce7-5655f6a3ba09

## About the Role

Toyota Financial Services (TFS) Technology team is looking for a highly motivated person to fill a role as an ML/AI Security Lead within the Security Intelligence Engineering organization. You'll own the intelligence layer of a new AI-powered security platform - starting with prompt engineering and managed AI service integration, then progressing to fine-tuning models on enterprise security data, and building a multi-model serving and routing layer. This role is what makes the organization own its intelligence rather than renting it from a vendor. You'll train models that understand the specific security environment, build the feedback loops that make them better over time, and ensure the AI layer delivers high accuracy on alert triage while keeping costs predictable through intelligent model routing., 3+ years in applied ML/AI engineering (not research-only - production deployment required)

Hands-on experience with LLM fine-tuning - LoRA, QLoRA, or full fine-tuning on domain-specific data

Experience with cloud ML platforms (e.g., AWS SageMaker): training jobs, hyperparameter tuning, model registry, endpoint deployment

PyTorch proficiency for model training and custom architecture

Experience building evaluation pipelines - automated metrics, human evaluation protocols, A/B testing

Understanding of transformer architectures and attention mechanisms (not just API calls)

Python fluency with production engineering practices (testing, CI/CD, monitoring)

Strong communication skills with the ability to explain model behavior and limitations to non-ML stakeholders

Added bonus if you have

Experience with security or cybersecurity data - alert classification, threat detection, anomaly detection

Familiarity with model serving at scale (vLLM, Triton Inference Server, Tensor RT optimization)

Hugging Face Ecosystem Experience Model Hub, tokenizers, datasets library, PEFT

Experience with RAG architecture and vector databases

Background in multi-model routing or mixture-of-experts approaches

Understanding of agentic AI patterns - tool use, chain-of-thought, multi-step reasoning

Experience with model cost optimization - quantization, distillation, caching strategies

Self-motivated individual who thrives in ambiguous environments and can build processes from the ground up

## Description

Design and implement prompt engineering patterns for managed AI service integration

Build training data pipelines from the security data lake - curating, labeling, and versioning datasets from real enterprise security telemetry

Fine-tune models on organization-specific security data - alert triage, risk scoring, finding classification

Implement the analyst feedback loop - capturing human corrections to continuously improve model accuracy

Build model evaluation frameworks with rigorous metrics (F1, precision, recall, false positive rates) benchmarked against analyst agreement

Design and implement a model routing layer - directing each task to the optimal model based on complexity, latency requirements, and cost

Monitor models in production for drift, accuracy degradation, and emerging failure modes

Implement centralized token usage monitoring for leadership visibility into AI consumption and cost control

Collaborate with the Lead Engineer on agent architectures - multi-agent orchestration, tool use, and autonomous triage workflows

Deploy and manage model inference endpoints across cloud ML services and container-based serving

Build the analyst feedback loop: approval/rejection signals in dashboards feeding back into retraining pipelines