> Markdown version of [/jobs/ext/4754428-ai-engineer](https://www.wearedevelopers.com/jobs/ext/4754428-ai-engineer). Every page supports `.md` or `Accept: text/markdown`. Links point to the HTML versions so they work for humans too. Agent guide: [/agents.md](https://www.wearedevelopers.com/agents.md). --- # AI Engineer - **Company:** Tata Consultancy Services Ltd - **Location:** Epsom, United Kingdom - **Contract:** Permanent contract - **Skills:** Clean Code Principles, Adobe InDesign, Application Programming Interfaces (APIs), Artificial Intelligence, Amazon Web Services, Business Analytics Applications, Audit Trail, User Authentication, Automation of Tests, Microsoft Azure, Cloud Engineering, Configuration Management, Code Reuse, Code Review, Databases, Continuous Integration, Data as a Services, Information Leak Prevention, DevOps, Distributed Computing Environment, Graph Database, Information Extraction, Python (Programming Language), Machine Learning, Metadata, Routing, Parsing, Regression Testing, Next.js, Software Engineering, SQL Databases, Data Streaming, Systems Integration, Management of Software Versions, Core Voice Platform, Data Logging, Google Cloud, Enterprise Software Applications, Cloud Platform System, ReactJS, Retrieval-Augmented Generation, Delivery Pipeline, Large Language Models, Indexer, Backend, Agentic-AI, Vue.js, Build Management, AI Platforms, AngularJS, Kubernetes, Information Technology, Low Latency, Deployment Automation, Enterprise Integration, Machine Learning Operations, Invoking Functions, Grpc, Software Version Control, Docker, Microservices - **Published:** November 17, 2026 - **Apply:** https://www.apply4u.co.uk/jobs/ai-engineer/56878869 ## About the Role software, data, ML or platform engineering experience, including hands-on ownership of production or production-like systems Hands-on delivery of LLM, retrieval, agentic or ML applications beyond tutorials or isolated notebooks Experience integrating models and AI services with enterprise applications, APIs, data sources, workflows and systems of record Experience with version control, automated testing, code review, containers, CI/CD, logging, monitoring and secure configuration Practical experience with at least one major cloud platform and its application, data, identity, networking, monitoring and deployment services Ability to work directly with users, architects and engineers and take a use case from technical specification through build, test, deployment and handover, Software & Integration Engineering Strong Python programming, API development and integration skills REST or gRPC, microservices, authentication, asynchronous or event-driven processing, databases and object storage Version control, automated testing, code review, CI/CD, Docker or similar containers and secure configuration Ability to understand unfamiliar code, isolate defects and explain implementation decisions Model APIs or enterprise model services, structured outputs, prompt and configuration management, tool or function calling and context handling Agent frameworks or equivalent orchestration approaches covering state, memory, retries, permissions and human checkpoints RAG and knowledge applications using embeddings, vector or hybrid search, metadata, ranking, access-aware retrieval and evaluation Understanding of model and agent failure modes, prompt injection, hallucination, data leakage and safe fallback patterns Data, Platform & Production Operations SQL and data-engineering skills for ingestion, transformation, data quality, metadata and schema evolution Cloud-native application and data services on AWS, Microsoft Azure or Google Cloud Kubernetes or managed container services, IaC, identity, networking, secrets, logging and monitoring MLOps or LLMOps practices covering versioning, automated evaluation, deployment, monitoring and rollback Ability to troubleshoot across data, retrieval, models, prompts, orchestration, integrations, infrastructure, performance and security Quality, Security & Professional Skills Creation of representative evaluation data, quality measures, thresholds, regression tests and release evidence Least-privilege access, masking or redaction, guardrails, audit evidence and secure tool execution Clear communication of progress, limitations, decisions and risks to technical and nontechnical stakeholders Ownership of end-to-end outcomes, with balanced attention to speed, maintainability, security, reliability and cost Knowledge sharing through pairing, reviews, demonstrations, documentation and reusable examples Preferred Qualifications Full-stack engineering experience with React, Angular, Vue, Next.js or comparable frameworks Experience with streaming, distributed processing, analytical platforms, knowledge graphs or document intelligence Experience in regulated, security-sensitive or high-availability environments Experience with conversational or voice AI, contact-center integration or multimodal applications Relevant cloud developer, AI engineer, data engineer, DevOps, Kubernetes or application-security certification Bachelor's or master's degree in Computer Science, Engineering, Data Science or a related field, or equivalent professional experience ## Description We are looking for a hands-on AI Engineer to build, integrate, test and deploy production-grade AI solutions. The engineer works with the AI Delivery Architect, client teams, domain teams and other delivery specialists to turn an approved design into working software. The role covers applied AI, software engineering, data and knowledge pipelines, enterprise integration, deployment automation, evaluation, observability and production troubleshooting. The role is important because a successful AI solution must work reliably beyond a demonstration. The AI Engineer owns the practical build, proves the solution through testing and telemetry, resolves technical issues and leaves behind reusable code, documentation and operational assets that the wider team can support and extend., Work with the AI Delivery Architect, client stakeholders and domain teams to convert approved requirements into technical specifications, tasks, acceptance criteria and working increments Build and deploy AI-enabled applications and services across back-end, APIs, orchestration and, where required, front-end components Develop AI agents and sub-agents using appropriate patterns for planning, routing, tools, memory, state, delegation, human approval and safe failure handling Build data, retrieval and context pipelines for structured and unstructured content, including ingestion, parsing, chunking, metadata, indexing, vector or hybrid search, ranking, filtering and evidence traceability Integrate model endpoints and gateways, and implement prompt, model and configuration versioning, routing, fallback and environment-specific settings Build and maintain data ingestion and transformation pipelines with validation, data-quality checks, metadata, lineage and schema-change handling Build governed integrations using REST or gRPC APIs, MCP, microservices, queues, events or streams, and enterprise systems Implement authentication, authorization, secrets, validation, idempotency, retries, rate limits, error handling, compensation and audit logging for AI-enabled actions Write clean, modular, testable, observable and maintainable code, supported by unit, integration, regression, security and performance tests Create evaluation datasets and automated tests for retrieval quality, groundedness, task success, policy adherence, tool selection, safety, latency, cost and end-to-end outcomes Instrument models, retrieval, agents, tools, integrations and applications with logs, metrics, traces, dashboards and alerts; use the resulting evidence to diagnose failuresBuild and maintain deployment pipelines, containers, infrastructure configuration, release controls, rollback procedures and operational runbooks Tune reliability, availability, throughput, latency and cost using caching, routing, context control, scaling and deterministic fallbacks where appropriate Contribute reusable code, agent templates, prompt and tool contracts, connectors, evaluation assets, reference implementations and engineering documentation Demonstrate working software, participate in design and code reviews, support deployment and troubleshooting, and transfer knowledge to the wider team, The engineer converts the approved design into stable, testable and observable working software Data, retrieval, model, integration and deployment components work together reliably in the target environment Evaluation and telemetry provide evidence for release decisions and ongoing improvement Reusable code and documentation reduce effort and risk for subsequent AI use cases The wider team can independently build, deploy, troubleshoot and extend the solution after the incubation phase