LLM Research Engineer

LLM Research Engineer

FocusKPI is looking for an LLM Research Engineer to join one of our clients, a high-tech SaaS company. Work Location: Mountain View, CA Duration: 12-month contract; Hybrid role (4 days per week onsite) Pay Range: $110/hr to $120/hr Role & Responsibilities:

• Design, train, and fine-tune large language models (e.g., GPT, LLaMA, PaLM) for various applications.
• Research cutting-edge techniques in natural language processing (NLP) and machine learning to improve model performance.
• Explore advancements in transformer architectures, multi-modal models, and emergent AI behaviors.
• Collect, clean, and preprocess large-scale text datasets from diverse sources.
• Develop and implement data augmentation techniques to improve training data quality.
• Ensure data is free from bias and aligned with ethical AI standards.
• Optimize model architecture to improve accuracy, efficiency, and scalability.
• Implement techniques to reduce latency, memory footprint, and inference time for real-time applications.
• Collaborate with MLOps teams to deploy LLMs into production environments using Docker, Kubernetes, and cloud.
• Develop robust evaluation pipelines to measure model performance using key metrics like accuracy, perplexity, BLEU, and F1 score.
• Continuously test for bias, fairness, and robustness of language models across diverse datasets.
• Conduct A/B testing to evaluate model improvements in real-world applications.
• Stay updated with the latest advancements in generative AI, transformers, and NLP research.
• Contribute to research papers, patents, and open-source projects—present findings and insights at conferences and internal knowledge-sharing sessions.

Qualifications:

• Recommend 7-10 years of experience
• Advanced degree in Computer Science, Artificial Intelligence, Data Science, or a related field.
• Strong programming skills.
• Proficiency with deep learning frameworks such as TensorFlow, PyTorch, or JAX.
• Hands-on experience with transformer-based models (e.g., GPT, BERT, RoBERTa, LLaMA).
• Expertise in natural language processing (NLP) and sequence-to-sequence models.
• Familiarity with Hugging Face libraries and OpenAI APIs.
• Experience with MLOps tools like Docker, Kubernetes, and CI/CD pipelines.
• Strong understanding of distributed computing and GPU acceleration using CUDA.
• Knowledge of reinforcement learning and RLHF (Reinforcement Learning with Human Feedback).

No C2C resumes are considered.