AI Backend Developer

Lorien

London, UK

13 days ago

Apply on www.careerboard.com

Prepare application

Role details

Contract type

Temporary contract

Employment type

Full-time (> 32 hours)

Working hours

Regular working hours

Job source

www.careerboard.com

Tech stack

.NET Framework Application Programming Interfaces (APIs) Artificial Intelligence Application Performance Management Microsoft Azure C Sharp (Programming Language) Continuous Integration Relational Databases Document-Oriented Databases Github Python (Programming Language) NoSQL

+27 more

OAuth Query Optimization Openid Connect Azure Active Directory Azure Data Lake Search Technologies Microsoft SharePoint Software Deployment Software Engineering Systems Integration Openapi Azure Service Bus Large Language Models Caching Indexer Backend Rate Limiting Event Driven Architecture Low Latency Bicep Cosmos DB Restful APIs Terraform Pagination Network Server Api Management Serverless Computing

Job description

We are currently recruiting for an AI Backend Developer with Azure AI experience to join one of our Insurance clients on a 6-month contract., * Design, build and maintain the Agent Data API (agent state, threads, conversation history, audit and feedback data in Azure Cosmos DB) as a secure, versioned, well-documented service.

Design, build and maintain the Member Data API as the controlled gateway to the BCUK system of record, including integration, mapping, caching, resilience and rate limiting.
Model data for Azure Cosmos DB with deliberate partitioning, indexing, consistency and TTL choices, and tune for cost and request-unit efficiency.
Build the ingestion and indexing pipelines that feed the knowledge base into Azure AI Search from SharePoint, Azure Data Lake and GitHub - extraction, normalisation, chunking, enrichment, embedding generation, incremental refresh and deletion handling.
Implement authentication and authorisation throughout: Entra managed identities for service-to-service calls, on-behalf-of and delegated permission flows where user context matters, and enforcement of document-level and record-level access so agents can never surface data a user is not entitled to see.
Expose clean, strongly typed API contracts (OpenAPI) suited to tool calling by agents, with predictable error semantics, idempotency and pagination.
Engineer for performance and reliability: async patterns, caching, connection management, retries with backoff, circuit breakers, and defined latency budgets that keep agent responses acceptable.
Write automated unit, integration and contract tests, and support the CI/CD pipeline through to production deployment.

Requirements

Strong Back End development experience in C#/.NET and/or Python, building and running RESTful APIs in production.
Solid Azure application development: App Service, Container Apps or Azure Functions, plus API Management or an equivalent gateway.
NoSQL and relational data modelling experience, with Azure Cosmos DB (or comparable document database) partitioning and query tuning.
Systems integration experience - consuming third-party or Legacy line-of-business APIs and handling their failure modes, latency and data quality issues.
Practical implementation of OAuth 2.0/OpenID Connect and Microsoft Entra ID authentication, including managed identity and service-to-service auth.
Observability instrumentation (Application Insights, OpenTelemetry or similar).
Experience building Azure AI Search indexes, indexers, skillsets or custom enrichment pipelines.
Experience implementing APIs specifically for LLM tool calling or MCP Servers.
Microsoft Graph/SharePoint Online and Azure Data Lake (ADLS Gen2) integration experience.
Event-driven architecture with Service Bus, Event Grid or Event Hubs.
Infrastructure-as-code familiarity (Bicep or Terraform) sufficient to deploy your own services.
Certifications such as AZ-204 or AI-102.