Senior Data Analysis & Evaluation for LLM Robustness

Deepstreamtech; Greater London; On-site; GBP 50,000 - 70,000; Full time

Job summary

Deepstreamtech is seeking a Member of Technical Staff in Data Analysis and Evaluation to ensure the quality and performance of large language models (LLMs). This role requires expertise in designing data collection tasks, assessing dataset quality, and working collaboratively with engineers and researchers. Ideal candidates should possess strong software engineering skills and hands-on experience with LLM training. The position offers an opportunity to contribute to advancing AI through robust and scalable systems.

Qualifications

• Extremely strong software engineering skills.
• Strong expertise in designing and conducting data collection tasks.
• Strong statistical skills and experience evaluating scientific experiments.
• Experience analysing datasets for quality and suitability for ML models.
• Hands-on experience training large language models on distributed infrastructures.

Responsibilities

• Ensure quality, reliability, and performance of large language models.
• Design and conduct data collection tasks.
• Assess and evaluate dataset quality.
• Work closely with cross-functional teams.
• Train and fine-tune large language models on distributed infrastructures.

Skills

Software engineering skills

Data collection task design

Statistical skills

Dataset analysis

Training large language models

Programming in Python

ML frameworks (e.g., PyTorch, TensorFlow, JAX)

Communication skills

Job description

Deepstreamtech is seeking a Member of Technical Staff in Data Analysis and Evaluation to ensure the quality and performance of large language models (LLMs). This role requires expertise in designing data collection tasks, assessing dataset quality, and working collaboratively with engineers and researchers. Ideal candidates should possess strong software engineering skills and hands-on experience with LLM training. The position offers an opportunity to contribute to advancing AI through robust and scalable systems.