AI/ML Engineer — AWS, LLMs, MLOps

VDart Inc; Auburn Hills (MI); On-site; USD 90,000 - 130,000; Part time

Job summary

VDart Inc. is seeking a Data Scientist to design, build, and operate ML/AI solutions on AWS, with a strong focus on LLMs, agentic AI, and RAG architectures. The role emphasizes governance, security, and end-to-end model lifecycle management in an onsite contract setting.

The candidate should have hands-on experience with Python/Java, containerization, and front-end delivery of model outputs, plus CI/CD and DevSecOps practices across ML workflows.

Qualifications

• Experience building ML solutions on AWS and deploying them to production.
• Hands-on expertise with LLMs and OpenAI APIs; knowledge of leading AI/LLM families.
• Proven ability to design agentic AI systems (multi-agent orchestration, tool use, autonomous workflows).

Responsibilities

• Design, build, train, and validate ML models with strong data understanding.
• Develop solutions using LLMs and generative AI, staying current with model releases.
• Build agentic AI solutions (multi-step, tool-using, autonomous agents).
• Create RAG architectures with embeddings and vector search; evaluate retrieval strategies.
• Implement HITL/HOTL patterns and governance in inference and retraining workflows.
• Provide operational support, monitoring, and incident triage for ML/AI services.
• Own model update lifecycle: retraining, fine-tuning, versioning, re-validation.

Skills

AWS

LLMs

Agentic AI

RAG

Python

Java

React/TypeScript UI

Docker

CI/CD

DevSecOps

Governance & Monitoring

HITL / HOTL

Testing & QA

Model Governance

Tools

SageMaker

OpenAI APIs

Vector Databases

Docker

CI/CD pipelines

PyTorch

TensorFlow

Hugging Face

LangChain

Job description

VDart Inc. is seeking a Data Scientist to design, build, and operate ML/AI solutions on AWS, with a strong focus on LLMs, agentic AI, and RAG architectures. The role emphasizes governance, security, and end-to-end model lifecycle management in an onsite contract setting.

The candidate should have hands-on experience with Python/Java, containerization, and front-end delivery of model outputs, plus CI/CD and DevSecOps practices across ML workflows.