Gen AI Engineer Models, Fine Tuning

Job Title: Gen AI Engineer - Model Fine-Tuning

Location: Dallas TX (3 days a week Hybrid)
Engagement Type: Contract

Overview

This is a hands-on role requiring deep expertise in LLM fine-tuning data curation and reinforcement learning optimization with the goal of reducing model hallucinations and enhancing contextual accuracy for production-grade cognitive systems.

Key Responsibilities

• Fine-tune large-scale LLMs (e.g. GPT Claude LLaMA Mistral) using curated domain datasets for banking risk and compliance workflows.
• Collaborate with data engineering teams to build high-quality labeled datasets for supervised and reinforcement learning.
• Apply advanced context engineering and prompt optimization techniques to improve model interpretability and reasoning.
• Evaluate and mitigate model drift bias and hallucination using quantitative performance metrics.
• Develop and automate evaluation pipelines for continuous fine-tuning and model retraining.
• Partner with the Cognitive Agent Development team to integrate tuned models into agentic workflows and decision chains.
• Contribute to model governance versioning and audit frameworks to ensure explainability and compliance.

Required Skills & Experience

• 5 10 years of hands-on experience in AI/ML with a focus on LLM fine-tuning prompt engineering or context adaptation.
• Strong proficiency with Python PyTorch TensorFlow and frameworks like Hugging Face Transformers LangChain and PEFT (Parameter-Efficient Fine-Tuning).
• Proven experience building and labeling domain-specific datasets and applying data augmentation strategies.
• Familiarity with RLHF (Reinforcement Learning with Human Feedback) and evaluation metrics for generative models.
• Understanding of multi-agent architectures orchestration frameworks (LangGraph CrewAI AutoGen etc.) and memory management for AI agents.
• Exposure to banking risk analytics or compliance data preferred.
• Strong grounding in data security privacy and model governance standards in regulated industries.

Preferred Qualifications

• Masters or PhD in Computer Science AI or related discipline.
• Experience deploying LLM-based agents in production environments.
• Knowledge of vector databases (FAISS Pinecone Chroma) and retrieval-augmented generation (RAG) pipelines.
• Contributions to open-source AI projects or publications in fine-tuning evaluation or multi-agent systems.