Member of Technical Staff – Applied Vision Post‑Training

About the role

Liquid‑AI is looking for a Member of Technical Staff to lead applied post‑training work on vision‑language models (VLMs). You will bridge cutting‑edge research and large‑scale enterprise deployments, ensuring that VLMs meet customer requirements and perform reliably in production.

Key responsibilities

• Own end‑to‑end VLM post‑training projects for enterprise customers, from requirement gathering to delivery and evaluation.
• Translate customer needs into concrete multimodal post‑training specifications and workflows.
• Design and execute visual data generation, filtering, and quality‑assessment pipelines, including image‑text pair curation, annotation, and synthetic data creation.
• Run supervised fine‑tuning, preference alignment, and reinforcement‑learning workflows for vision‑language models.
• Develop task‑specific evaluations for visual understanding, grounding, OCR, document parsing and other multimodal capabilities, interpret results and feed improvements back into core pipelines.

Required profile

• Demonstrates ownership and can drive VLM post‑training projects from start to finish.
• Think end‑to‑end across data curation, model training, alignment and evaluation.
• Pragmatic focus on model quality and customer outcomes over academic publications.
• Clear communicator who can bridge customer expectations and internal technical teams.

Required skills

• Hands‑on experience with data generation and evaluation for vision‑language models or multimodal post‑training.
• Experience training or fine‑tuning VLMs using supervised fine‑tuning, preference alignment, and/or reinforcement learning.
• Strong intuition for visual data quality, annotation design, and multimodal evaluation.
• Familiarity with vision encoders, image‑text architectures and their interaction with language model backbones.