Ava Developer
Role details
Job location
Tech stack
Job description
Software Engineering & Delivery:
- Design, Build, and ship production quality services within our current business controls
- Build and Optimise DecOps pipelines - enhancing CI/CD processing, including testing and platform/process observability
- Design, iterate, and optimise prompts for a wide range of LLM use cases - instruction following, structured output generation, classification, summarisation, code generation, and reasoning tasks
- Build and manage prompt libraries and versioning strategies, treating prompts as first-class engineering artefacts
- Collaborate with product and domain teams to translate business requirements into precise, well-structured prompt specifications
Agentic AI & LLM Integration
- Architect and implement/maintain agentic systems - autonomous agents, multi-agent pipelines, tool-use workflows, planning loops, and memory-augmented reasoning
- Implement LLM observability - logging, tracing, evaluation pipelines, and guardrails - to maintain quality and reliability in production AI systems
- Advise on responsible AI deployment, including output validation, human-in-the-loop design, and risk mitigation for agentic workflows
Technical Skills & Experience
Full-Stack Engineering
Requirements
Strong proficiency in Java/Spring Boot - REST APIs, event-driven services, security, and performance optimisation
Solid React development skills and API integration
Experience with relational and in memory databases (PostgreSQL, Redis)
Messaging and streaming platforms - Kafka, or equivalent
Prompt Engineering
Demonstrable, hands-on Experience engineering prompts for production LLM applications - not just prototypes
Familiarity with prompt evaluation frameworks and tooling (PromptFlow, LangSmith, Braintrust, or custom eval pipelines)
Understanding of tokenisation, context limits, temperature tuning, and model-specific behaviours across major LLM providers
Experience structuring prompts for tool/function calling, structured JSON output, and multi-turn conversations
DevOps & Infrastructure
Kubernetes - deployments, services, ingress, Helm, Kustomize, HPA, and RBAC
CI/CD pipelines - GitHub Actions, GitLab CI, Jenkins, or equivalent
Infrastructure as code - Terraform
Cloud platforms - AWS, GCP, or Azure