Job Description
We are looking for an Applied AI & Enterprise Systems designer, develops, tests and supports enterprise software incorporating Generative AI capabilities. The role combines full-stack development, backend services, data integration and cloud deployment with AI model integration, retrieval-augmented generation (RAG), agentic workflows and evaluation.
The engineer will work with project managers, business stakeholders and technical specialists to take solutions from proof of concept through pilot deployment and production support.
Scope of work
1. Software design and full-stack development
- Analyse business and technical requirements and translate them into application designs, development tasks and software components.
- Develop backend services using Python, FastAPI, C#/.NET, Node.js or other technologies appropriate to the project.
- Build responsive web interfaces using React, Next.js, TypeScript or equivalent frameworks.
- Develop RESTful APIs and integrate applications with enterprise systems, databases and external services.
- Implement streaming responses and asynchronous processing where required.
- Apply validation, exception handling, logging, timeouts and retry mechanisms to support reliable application behaviour.
2. Generative AI application development
- Integrate approved foundation models through cloud services, model APIs or supported inference platforms.
- Develop components for prompt construction, model invocation, structured outputs and tool calling.
- Build AI assistants and knowledge applications that provide access to enterprise information.
- Evaluate model configurations against application requirements, including output quality, response time and operating cost.
- Implement input validation, content controls, usage tracking and other safeguards required by the application.
3. Data ingestion and retrieval
- Build pipelines to ingest and process documents, files, APIs and database records.
- Implement document parsing, data validation, chunking, embedding generation and indexing.
- Develop RAG workflows using suitable orchestration frameworks and data stores.
- Implement and evaluate keyword, vector and hybrid retrieval, including query expansion and reranking where appropriate.
- Maintain source references and enforce access permissions when retrieving enterprise information.
- Investigate ingestion failures, retrieval errors and data quality issues.
4. Agentic workflows and tool integration
- Develop multi-step AI workflows using frameworks such as LangGraph or equivalent tools.
- Integrate agents with approved APIs, databases and enterprise tools, including Model Context Protocol (MCP) interfaces where appropriate.
- Implement workflow state, conversation persistence and controlled handoffs between application components or agents.
- Apply authentication, scoped tool access and audit logging.
- Test workflow behaviour, tool execution and recovery from failed or incomplete operations.
5. Experimentation, testing and evaluation
- Define evaluation objectives and representative test cases with business stakeholders and subject matter experts.
- Compare models, prompts and retrieval strategies through repeatable experiments.
- Assess answer accuracy, relevance, groundedness, retrieval quality, latency and cost.
- Investigate hallucinations, factual errors and inconsistent responses and implement corrective changes.
- Develop unit, integration, end-to-end and prompt regression tests.
- Document experiment configurations, results and recommendations to support delivery decisions.
6. Cloud deployment and production support
- Package and deploy applications using containers and approved cloud infrastructure.
- Maintain Git-based source control and CI/CD pipelines for automated testing and deployment.
- Implement health checks, structured logging, tracing, monitoring and alerts.
- Monitor application performance, model usage, response quality and operating costs.
- Support incident investigation, defect resolution, application enhancements and release rollback.
- Prepare deployment guides, operational runbooks and technical handover documentation.
7. Technical leadership and collaboration
- Contribute to architecture decisions and explain technical options, dependencies and trade-offs.
- Estimate development effort and coordinate implementation with local and offshore engineering teams.
- Participate in code reviews and guide developers on implementation and testing.
- Work with data scientists and AI specialists on evaluation and production readiness.
- Communicate technical findings and delivery progress to project managers and stakeholders.
- Support knowledge transfer and ongoing maintenance of delivered applications.
Expected deliverables
- Application designs and documented integration interfaces.
- Working prototypes, pilot releases and production application components.
- AI-enabled user interfaces, backend services and APIs.
- Data ingestion, retrieval and agentic workflow components.
- Automated tests, evaluation results and defect resolutions.
- Deployment pipelines, monitoring configurations and operational documentation.
Requirements
- A qualification in Computer Science, Information Systems, Software Engineering or a related discipline, or equivalent relevant practical experience.
- Demonstrated experience delivering enterprise software across design, development, testing, deployment and support.
- Strong backend programming skills in Python, C#/.NET, Node.js or a comparable technology.
- Experience developing web applications, APIs and database integrations.
- Practical experience integrating Generative AI models into applications.
- Understanding of prompting, embeddings, RAG, tool calling and model evaluation.
- Experience with relational databases and familiarity with search or vector data stores.
- Familiarity with cloud deployment, containers, Git, CI/CD and automated testing.
- Ability to investigate technical issues and communicate findings clearly.
- Experience owning application components, reviewing code or guiding other developers.
Preferred experience
- FastAPI, React, Next.js and TypeScript.
- LangGraph, LangChain, LlamaIndex or equivalent frameworks.
- MCP integration and multi-agent workflows.
- Evaluation and observability tools such as DeepEval, Ragas, LangSmith or Langfuse.
- AWS or Google Cloud, including managed container and AI services.
- Hybrid search, reranking, streaming interfaces and background processing.
- Application security, authentication, audit logging and production incident support.
You apply on MyCareersFuture, where this job was posted. Peter Nowak isn’t the employer and doesn’t check listings.
