Senior LLM Engineer - Production-Grade Fine-Tuning (Remote)

Bright Vision Technologies; Woodbridge Township (NJ); Remote; USD 100.000 - 150.000; Vollzeit

Zusammenfassung

Bright Vision Technologies is seeking an LLM Engineer to design, execute, and operationalize fine-tuning workflows for large language models across supervised, preference-based, and reinforcement learning approaches.

This role requires deep practical experience with modern training stacks, careful dataset construction, rigorous evaluation methodology, and the engineering discipline to operate complex training pipelines reliably.

Qualifikationen

• Master’s or PhD in CS/ML or related field, or equivalent experience.
• 10+ years combined ML research and engineering with LLM exposure.
• Strong Python and PyTorch proficiency.
• Hands-on fine-tuning of transformer models at scale.
• Familiarity with distributed training (FSDP, ZeRO, pipeline parallelism).
• Experience with RLHF, DPO, or similar preference methods.
• Strong evaluation methodology and human evaluation design.
• Experience operating GPU training jobs and recovering from failures.
• Excellent written and verbal communication skills.
• Track record of shipping impactful LLM work.

Aufgaben

• Design and execute fine-tuning experiments for LLMs.
• Lead dataset construction and QA for instruction tuning.
• Build scalable training pipelines on modern frameworks.
• Tune hyperparameters and stability for large models.
• Implement LoRA/QLoRA and adapter-based methods.
• Design rigorous evaluation suites and benchmarks.
• Implement safety and policy evaluations across releases.
• Operate large-scale GPU training jobs and recover state.
• Mentor engineers and share best practices.

Kenntnisse

Python

PyTorch

LLM basics

Distributed training

RLHF

DPO

Evaluation design

Communication

Ausbildung

Master’s/PhD in CS/ML or related

Tools

FSDP

ZeRO

Pipeline parallelism

Jobbeschreibung

Bright Vision Technologies is seeking an LLM Engineer to design, execute, and operationalize fine-tuning workflows for large language models across supervised, preference-based, and reinforcement learning approaches.

This role requires deep practical experience with modern training stacks, careful dataset construction, rigorous evaluation methodology, and the engineering discipline to operate complex training pipelines reliably.