Role overview
- Design, train, and fine-tune LLMs and domain-specific NLP models for text understanding, summarization, and retrieval tasks
- Develop scalable data pipelines, embeddings, and vector search systems for unstructured construction data
- Integrate RAG (Retrieval-Augmented Generation) systems into production applications
- Collaborate cross-functionally with product and platform teams to deliver ML-driven product features end-to-end
- Evaluate and optimize model performance for latency, accuracy, and robustness in production environments
Basic qualifications
- 4+ years of professional industry experience, with at least 2 years focused on NLP
- Proven experience training and fine-tuning LLMs and transformer-based architectures (e.g., GPT, BERT, T5, LLaMA)
- Proficiency in NLP frameworks and libraries such as Hugging Face Transformers, spaCy, PyTorch, TensorFlow, LangChain, and OpenAI APIs
- Experience with search and retrieval technologies including Elasticsearch, Pinecone, Milvus, Weaviate, FAISS, or Vespa
- Familiarity with RAG architectures and building context-aware AI systems
- Strong programming skills in Python and experience deploying ML models in cloud environments (AWS, GCP, or Azure)
- Prior experience in startups or small, fast-paced teams with ownership of full product lifecycles
Benefits
- Competitive cash compensation and meaningful equity stake
- Health, dental, and vision insurance coverage
- Flexible hybrid and remote work options
- Unlimited PTO
Tags & focus areas
Used for matching and alerts on DevFound Ai Machine Learning Nlp Generative Ai Pytorch Tensorflow Fulltime