Location: hybrid/remote Reports to: CTO (permanent role) - No Visa Sponsorship The company An early-stage legal tech SaaS business. Its platform automates document-heavy legal workflows for law firms and institutional clients. It has product-market fit, a growing customer base and a lean engineering team. The role You will lead the development of AI features in the core product. This is a hands-on role. - Design and build ML models for document parsing, search, classification and summarisation - Integrate LLMs into product workflows with guardrails, evaluation and monitoring - Optimise model serving and inference for latency and cost on AWS (ECS, Lambda, S3) - Build data pipelines for training and evaluation (PostgreSQL, Redis, OpenSearch) - Work with engineering, product and QA in Scrum squads, with CI/CD and infrastructure as code already in place Requirements - 5+ years as an AI/ML engineer delivering NLP and ML models into production SaaS - Production LLM experience: RAG, fine-tuning, evaluation, monitoring - Strong Python, including building APIs and integrating models into applications - NLP frameworks such as Hugging Face Transformers, PyTorch, spaCy or TensorFlow - AWS and distributed systems - Data pipelines and orchestration for model training and evaluation Nice to have - Legal tech or another document-heavy domain - Vector databases and semantic search (OpenSearch, Pinecone, FAISS) - Explainability and compliance work on deployed AI - Knows React or TypeScript ..... full job details .....