LLM AI Engineer

Description

Propio is on a mission to make communication accessible to everyone. As a leader in real-time interpretation and multilingual language services, we connect people with the information they need across language, culture, and modality. We’re committed to building AI-powered tools to enhance interpreter workflows, automate multilingual insights, and scale communication quality across industries. 

We are seeking an LLM AI Engineer to join our AI team and help build cutting-edge products using large language models (LLMs). You’ll work closely with a team of AI engineers and the product team to build systems that summarize, translate, evaluate, and structure multilingual interpretation content in real time.

This is a high-impact role at the intersection of NLP, multilingual AI, and real-world healthcare/legal communication.

Key Responsibilities:

•  Fine-tune, evaluate, and deploy LLMs (e.g., Mistral, GPT, Claude, LLaMA) for summarization, QA, and multilingual tasks
• Build and optimize embedding pipelines, RAG systems, and vector search capabilities for document/question retrieval
• Work with Prompt Engineers to prototype use cases before transitioning to model-level customization
• Develop evaluation frameworks for language generation quality, semantic preservation, and multilingual fluency
• Collaborate with AI Product Manager to translate business requirements into scalable model solutions
• Work with MLOps to ship models into production with robust monitoring and fallback mechanisms
• Stay up to date with open-source developments and help evaluate new foundation models and libraries

Requirements

Qualifications:

•  Master’s Degree in Engineering, preferably in Computer Science, Statistics, Data Science or equivalent work experience
• 3+ years of experience working with NLP or large-scale ML models in production
• Strong understanding of transformer architectures and recent LLM developments (e.g., LoRA, RAG, PEFT)
• Hands-on experience with one or more frameworks: Hugging Face, LangChain, OpenAI APIs, DeepSpeed, or vLLM
• Proven experience building multilingual or domain-adapted LLM pipelines
• Familiarity with retrieval-augmented generation (RAG), vector stores (FAISS, Weaviate), and prompt evaluation
• Strong Python skills and comfort working in Git-based, containerized environments
• Experience with healthcare or regulated domains a plus (e.g., HIPAA, PHI-safe modeling)

Preferred Qualifications:

• Experience training or fine-tuning open-source LLMs from scratch
• Prior work in translation, interpretation, summarization, or cross-lingual QA
• Contributions to open-source NLP projects
• Experience working with speech-to-text pipelines or multimodal systems