AI Engineer III

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 hiring an AI Engineer III to build and deploy practical, high-impact AI systems. You will work across speech recognition, large language models, and prompt engineering to ship products like AI-generated summaries, interpreter QA tools, and multilingual retrieval systems. You’ll operate at the intersection of engineering, research, and product with high ownership and startup-level pace., * Prototype, build, and deploy end-to-end AI applications involving speech, LLMs, and text generation

• Integrate APIs like OpenAI, Whisper, Deepgram, and open-source equivalents for ASR and NLP
• Collaborate with Engineers to iterate and refine MVPs before transitioning to model-level optimization
• Develop internal tools and dashboards to test summarization, QA scoring, and multilingual understanding
• Rapidly test ideas and model variations to explore feasibility and impact (build-measure-learn loop)
• Ensure model pipelines are robust, scalable, and ready for handoff to MLOps and production
• Work with the AI PM to align technical outputs with business use cases and feedback loops
• Stay current on applied AI trends in speech and LLMs, and advise on what to use vs. build

• Master’s Degree in Engineering, preferably in Computer Science, Statistics, Data Science or equivalent work related experience
• 5+ years of experience working with NLP or speech models in real-world applications
• Experience with Python, Hugging Face Transformers, OpenAI APIs, Whisper, LangChain, or similar frameworks
• Experience deploying AI models in production or pilot environments (e.g., using FastAPI, Flask, or Streamlit)
• Strong understanding of embeddings, prompt chaining, and pipeline orchestration
• Familiarity with vector databases (e.g., FAISS, Pinecone, Weaviate)
• Comfortable with rapid iteration, MVP mindset, and cross-functional collaboration
• Prior exposure to multilingual or low-resource language challenges is a plus, * Experience building speech-to-text pipelines or hybrid ASR + LLM systems
• Experience with data labeling, annotation, or active learning workflows
• Familiarity with real-time audio processing or latency-sensitive applications
• Experience in healthcare, legal, or regulated environments (HIPAA, PHI, Section 1557)