Full-stack Language Model Engineer

Role Overview

As an AI Language Model Engineer, you will work on the development and optimization of Large Language Models (LLMs), Vision-Language Models (VLMs), and multimodal reasoning systems. You will work closely with our computer vision, robotics, embedded, and semiconductor teams to develop models capable of understanding visual information, reasoning about physical environments, generating structured outputs, and controlling downstream robotic or software systems.

Responsibilities

• Develop and improve O-HIVE's LLM and VLM architectures for industrial and robotic applications.
• Fine-tune and adapt open-source language and multimodal models using techniques such as SFT, LoRA, QLoRA, and PEFT.
• Develop multimodal pipelines combining images, video, natural language, and spatial information.
• Optimize models for low-latency edge inference on GPUs and embedded AI devices.
• Integrate language models with computer vision, detection, and robotic-control pipelines.
• Collaborate with hardware engineers to define model requirements for future VLM accelerator chips.

Requirements

• Bachelor's, Master's, or Ph.D. degree in Computer Science, AI, ML, or a related field.
• Strong programming skills in Python.
• Strong experience with PyTorch and modern deep-learning frameworks.
• Hands-on experience with Transformer-based models and architectures like Llama, Mistral, or Gemma.
• Experience with Hugging Face Transformers and ML tooling.
• Deep understanding of attention mechanisms, tokenization, and quantization.

Skills

• Python
• PyTorch
• LLMs
• Hugging Face
• Computer Vision