Senior LLM Engineer (Fine-Tuning & LLMOps)

• AI/ML

• Remote

• Contract

Title – Senior LLM Engineer (Fine-Tuning & LLMOps)
Start date: Immediate
Position Type: Contract
Location: Remote across Canada/USA
  
  
About the Role: 
We are looking for a highly skilled and experienced Senior LLM Engineer to lead the fine-tuning and deployment of large language models (LLMs) in local environments. This role involves working on end-to-end LLM workflows, including model customization, deployment, and lifecycle management using modern LLMOps practices. 
As a core contributor, you will develop robust CI/CD/CL (Continuous Integration/Continuous Deployment/Continuous Learning) pipelines, ensuring the efficient and scalable use of LLMs for various applications. 
  
Key Responsibilities: 
  
LLM Fine-Tuning: 

• Fine-tune pre-trained large language models (e.g., GPT, LLaMA, Falcon) for specific use cases and domain-specific tasks. 
• Design and implement custom data preprocessing and augmentation pipelines to improve model performance. 

  Local LLM Deployment: 

• Deploy and optimize LLMs on on-premise environments, ensuring resource efficiency. 
• Configure and manage GPU clusters for high-performance local inference and training. 

  LLMOps & Automation: 

• Develop robust CI/CD pipelines to automate model versioning, testing, and deployment. 
• Implement Continuous Learning (CL) pipelines to retrain models with fresh data and monitor performance. 

  Model Monitoring & Optimization: 

• Establish comprehensive monitoring systems for tracking model performance, drift, and latency. 
• Optimize inference workflows for low-latency and high-throughput performance. 

  Collaboration and Documentation: 

• Collaborate with cross-functional teams, including data scientists, DevOps engineers, and software developers, to integrate LLMs into production systems. 
• Maintain detailed documentation of workflows, pipelines, and model changes.  

Research and Development: 

• Stay updated with the latest advancements in LLMs, fine-tuning techniques, and LLMOps tools. 
• Experiment with new architectures and methodologies to improve efficiency and scalability. 

Key Requirements: 
  
Experience: 

• 7+ years of programming experience in machine learning or AI-related roles. 
• At least 3 years of hands-on experience fine-tuning and deploying large language models. 

  
Technical Skills: 

LLM Expertise: 

• Proficiency in fine-tuning frameworks like Hugging Face Transformers, LoRA (Low-Rank Adaptation), or PEFT (Parameter Efficient Fine-Tuning). 
• Experience with pre-trained models such as GPT, BERT, T5, LLaMA, or Falcon. 
• Knowledge of prompt engineering and evaluation techniques.   

Deployment: 

• Strong experience in deploying LLMs locally on GPUs and optimizing for performance. 
• Familiarity with tools like NVIDIA Triton Inference Server, TensorRT, and DeepSpeed.   

LLMOps: 

• Proficiency in tools like MLflow, DVC, or Kubeflow for model lifecycle management. 
• Expertise in CI/CD tools (e.g., GitLab Actions, Jenkins) and integrating them with machine learning pipelines. 
• Experience implementing Continuous Learning pipelines for model retraining.  

Programming Skills: 

• Strong Python programming skills with experience in libraries such as PyTorch and TensorFlow. 
• Proficiency in scripting and automation (e.g., Bash, Terraform).  

Infrastructure Management: 

• Experience with Kubernetes, Docker, and managing GPU clusters. 
• Familiarity with hybrid cloud and on-premise deployments.  

Preferred Qualifications: 

• Familiarity with distributed training frameworks (e.g., Horovod, PyTorch DDP). 
• Experience working with knowledge bases, RAG (Retrieval-Augmented Generation), or hybrid AI systems. 
• Strong understanding of secure deployment practices for sensitive applications.   

Soft Skills: 

• Excellent problem-solving skills and attention to detail. 
• Strong collaboration and communication abilities. 
• Self-motivated with a proactive approach to learning and experimentation.