AI Architect

AI Architect Experience: 10 to 15 years, Location: New Delhi (Turkman Gate), Work From Office About the Company Our client is an AI first product company that is revolutionising the legal, tax and accounting sectors in India with an advanced Gen AI platform. Unlike general AI tools, this platform is carefully designed and built by industry experts to meet the specific needs of the legal domain. The AI system helps professionals make better decisions by providing accurate and updated information and saves time by quickly analysing queries and offering practical insights. Role Overview We are hiring a senior AI Architect on behalf of our client. The selected candidate will own the complete architecture and delivery of a Gen AI enabled legal platform. This is a senior technical leadership role, similar in impact to a CTO, with a strong focus on building and scaling AI first systems. The AI Architect will define the AI, data and cloud architecture for a legal domain focused product that uses large language models, retrieval augmented generation and vector databases. The role involves working directly with the founders, senior engineers, product managers and legal experts to convert business and domain needs into reliable, secure and scalable AI solutions. Key Responsibilities Architecture and System Design Own the end to end technical architecture for the Gen AI platform, from data ingestion to model serving and integration with applications. Design scalable, modular and secure systems on cloud to support multiple legal, tax and accounting use cases. Select appropriate tools, frameworks and services for LLMs, RAG, vector databases, orchestration and monitoring. GenAI, LLM and RAG Solutions Architect and guide implementation of LLM based solutions for legal research, document summarisation, drafting and advisory workflows. Design retrieval augmented generation flows that combine LLMs with trusted legal and tax content to improve accuracy and reduce hallucinations. Define approaches for fine tuning, prompt design, evaluation and optimisation of models for legal domain tasks. Data, Vector Databases and Knowledge Layer Design data pipelines to ingest, clean, split and enrich legal, tax and accounting documents from different sources. Architect vector database or embedding store solutions to support high quality semantic search and retrieval across large legal corpora. Set standards for data quality, metadata, versioning and knowledge governance. Cloud, Security and Deployment Architect cloud infrastructure on AWS, Azure or GCP for hosting AI services, APIs and applications in a scalable and cost effective way. Ensure strong security practices for sensitive legal data, including access control, encryption, logging and audit readiness. Establish CI or CD and MLOps practices for automated deployment, monitoring and rollback of models and services. Product and Stakeholder Collaboration Work closely with product management and founders to translate product vision and client requirements into technical designs and delivery plans. Collaborate with legal domain experts so that AI solutions align with real world legal workflows and risk expectations. Break down complex initiatives into clear technical tasks and milestones for the engineering team. Technical Leadership and Team Development Provide technical direction, design reviews and guidance to GenAI engineers, data engineers and backend developers. Drive engineering best practices for code quality, documentation, testing and reliability. Support hiring and mentoring to build a strong AI and engineering team aligned with the defined architecture. Quality, Compliance and Continuous Improvement Define and track metrics for model quality, hallucination rates, relevance, latency and system reliability. Ensure adherence to applicable data privacy and regulatory guidelines relevant for legal and financial information. Continuously improve architecture, performance, cost efficiency and developer productivity. Qualifications Must Have 10 to 15 years of experience in software engineering and architecture, with significant exposure to AI, ML or data intensive systems. Proven experience in architecting and delivering production grade AI or ML solutions, preferably with NLP or LLM components. Strong programming skills in Python and experience with ML frameworks such as PyTorch or TensorFlow. Good understanding of LLM concepts, embeddings, prompt engineering, RAG architectures and vector databases. Hands on experience with at least one major cloud platform such as AWS, Azure or Google Cloud. Strong background in distributed systems, microservices and API based architectures. Experience in leading technical teams, driving architectural decisions and taking ownership for delivery in complex projects. Clear communication skills and ability to work with product, business and domain stakeholders. Good to Have Experience in legal tech, fintech, regtech or other regulated domains. Prior work on enterprise search, RAG systems or large document based solutions. Exposure to MLOps tools such as MLflow, Kubeflow, SageMaker, Vertex AI or similar. Experience in early stage startups where you have owned architecture and technology strategy. What Success Looks Like In 12 To 24 Months The ideal candidate should be confident of delivering most of the following outcomes for our client. Within 12 to 24 months, the AI Architect is expected to: Architect, build and take to production a Gen AI based legal platform that supports key legal, tax and accounting use cases for clients. Deliver a robust knowledge and vector database layer that can ingest and serve large volumes of legal documents with high retrieval accuracy. Implement a reliable RAG based architecture that clearly improves trust and reduces hallucinations in AI generated legal outputs. Set up secure and scalable cloud infrastructure with MLOps workflows so that models and services can be deployed, monitored and updated safely. Define and track clear system and model performance metrics and drive regular improvements based on those metrics. Build and mentor an effective AI and engineering team that can execute independently within the defined architecture. Translate business and legal requirements into a clear technical roadmap and consistently deliver against agreed timelines. Contribute to measurable business impact, such as faster query resolution, better accuracy of insights and higher adoption of AI features by users. Candidates should apply only if they have strong hands on experience in these areas and are comfortable taking complete ownership for architecture and delivery of an AI first product. Why This Opportunity Work on an AI first product where technical decisions directly shape how legal, tax and accounting work is delivered in India. Partner closely with founders, AI engineers, product managers and legal experts in a focused and collaborative environment. Take up a role with CTO like ownership while continuing to be closely involved in architecture and problem solving. Be part of a fast growing startup that is building advanced Gen AI solutions specifically for the legal, tax and accounting sectors.