Artificial Intelligence Developer

Overview :The AI Developer will design, build, test, deploy and support in-house AI agents for a UK legal firm. The role will focus on creating secure, reliable and auditable AI solutions that improve legal operations, document workflows, knowledge retrieval, client service processes and internal productivity while respecting confidentiality, data protection and professional standards relevant to legal practice.Key Responsibilities :- AI agent development : Design and develop task-focused AI agents using modern large language model frameworks, retrieval-augmented generation, workflow orchestration and API integrations.- Legal workflow automation : Build solutions for legal document review, summarisation, drafting assistance, matter triage, knowledge search, case workflow support, compliance checks and internal process automation.- Retrieval and knowledge systems : Create secure retrieval pipelines over internal legal knowledge bases, precedents, policies, templates and matter-related data using vector databases, embeddings, metadata filters and permissions-aware access control.- System integration : Integrate AI services with document management systems, case management systems, SharePoint, Teams, CRMs and internal databases through APIs and secure connectors.- Prompt and evaluation engineering : Develop prompt templates, agent instructions, guardrails, test sets, evaluation criteria and human-in-the-loop review processes to improve consistency, accuracy and safety.- Data security and compliance : Implement solutions that protect client confidentiality, personal data and privileged information, including logging controls, secure authentication, redaction workflows and appropriate data retention practices.- Production engineering : Deploy and monitor AI solutions in a reliable production environment, including CI/CD, version control, observability, error handling, performance optimisation and incident support.- Stakeholder collaboration : Work closely with legal, operations, compliance, IT security and marketing teams to understand requirements and translate them into practical technical solutions.- Documentation and training : Prepare technical documentation, user guides, release notes and internal training materials for AI tools and workflows.- Continuous improvement : Monitor user feedback, model performance, cost, latency and adoption metrics, and recommend improvements to the AI agent roadmap.Essential Skills and Experience :- Strong software development experience in Python and/or TypeScript/JavaScript.- Hands-on experience building applications with LLM APIs such as OpenAI, Azure OpenAI, Anthropic, Google Gemini or comparable enterprise AI platforms.- Experience with agentic AI frameworks or orchestration tools such as LangChain, LangGraph, LlamaIndex, Semantic Kernel, CrewAI or equivalent.- Practical understanding of retrieval-augmented generation, embeddings, vector databases and semantic search.- Experience designing REST APIs, background workers, queues and integrations with third-party systems.- Knowledge of cloud platforms, preferably Microsoft Azure, including identity, storage, functions/app services and monitoring.- Good understanding of secure software engineering, role-based access control, secrets management and audit logging.- Ability to write clean, maintainable, tested code using Git-based workflows.- Strong problem-solving skills and ability to work with non-technical stakeholders.- Excellent written and spoken English communication skills.Desirable Skills :- Experience working with document automation or e-discovery platforms.- Experience with Microsoft Graph, SharePoint, Teams, Power Platform or Dynamics integrations.- Experience with model evaluation, red-teaming, prompt injection mitigation and AI safety controls.- Familiarity with OCR, document parsing, PDF extraction, legal document classification and structured data extraction.- Experience containerising applications using Docker and deploying through CI/CD pipelines.- Understanding of data protection principles relevant to UK GDPR and confidential client information. (ref:hirist.tech)