Application Support Engineer

The Role

As an Application Support Engineer on the AI Platform, you'll keep the foundation Aflac runs its agentic AI on healthy across environments : an AWS native, secure, highly scalable platform that other engineering and product teams build on top of. Orchestration, tools, memory, identity, guardrails, evaluation and observability. You'll provide Level 2 and Level 3 support across the platform and the use cases that run on it and help make it production ready from the start rather than after go live.

Agentic systems are non - deterministic, so the support bar is higher than it looks. Anyone can make an agent work once. You'll help keep it working reliably, observably, safely , and at a cost the business can defend. That means treating monitoring, tracing, runbooks and incident response as first class engineering, not afterthoughts. In line with our "Prevention is better than the Cure" approach, you'll get to the root cause and put measures in place that stop issues coming back.

Serving as a technical owner, you'll use AI as a force multiplier and human amplifier to triage faster, automate routine work , and continually improve the production environment. You'll also be a trusted partner to the teams that use the platform, translating technical issues into plain language, and bringing what you learn in production back into the roadmap.

This is an opportunity to

Shape how an enterprise supports agentic AI in production

Build the operational controls, runbooks and monitoring while the decisions are still open

Work hands on with the newest AWS AI services at production scale

Solve genuinely hard problems in the reliability, observability and cost of agentic systems

Grow from application support into AI platform and site reliability engineering

Build something from the ground up and leave a lasting impact

About the Team

The AI Services team helps transform how Aflac uses AI at scale where it matters. Taking priority use cases from concept to production, proving what works early, and packaging the results so other teams can repeat them. This team leads Aflac's AI harness engineering methodology, equipping engineers and product teams with AI driven workflows and setting SDLC standards that make AI available, safe and scalable.

The Agentic Platform is the enterprise foundation for that ambition, and we are building it now from first principles. Agent orchestration, tool and MCP server registries, memory and state, identity and access, guardrails, tracing, cost control and self service onboarding.

AI Services Engineering runs as two functions. This role sits in the AI Platform function, owning the Agentic Platform: how Aflac runs agents at scale, cost efficiently and securely, with the right risk posture and robust operational controls. The other function is Context Engineering, taking AI native products from concept to production and setting the AI harness engineering methodology behind them. Alongside both sits AI Enablement, driving adoption, education and communication across the business. Two sides of one engineering team, working closely with the full AI Services team. You'll work with Platform Engineers to make the platform production ready and keep it that way, and with Context Engineering, Solution Architects, Enterprise Architecture, Security and QA in an agile environment.

Our teams take full product ownership from design through to deployment, with a proactive support model that includes incident management and a rotational on-call function. Core support hours are UK based, from 9am to 5pm, with on-call coverage as required.

What you'll be doing

Production Support

Provide Level 2 and Level 3 support for the Agentic Platform and the AI use cases that run on it, including triage, investigation and resolution

Operate in a core hour support model from 9am to 5pm UK time, and take part in an on-call rota for out of hours and weekend support

Lead problem management and blameless post incident reviews, and turn findings into lasting fixes

Support teams onboarding to the platform, and feed support insight back into the roadmap

Work with internal teams and third party vendors to troubleshoot and resolve complex problems

Reliability and Observability

Define and track service level objectives for availability, latency, quality and cost

Build and maintain monitoring, tracing and alerting across agents, models, tools, MCP servers and AWS infrastructure

Monitor AI specific signals, including evaluation scores, quality drift, guardrail triggers, tool failures and token usage

Track the operational impact of model and prompt changes, and support safe rollout and rollback

Support operational resilience across a multi region AWS environment, including disaster recovery testing

Automate routine operational work and runbooks, using AI driven tools where they help

Operational Controls and Risk

Help design and run the platform's operational controls, including human in the loop checkpoints, rate limits, kill switches and fallbacks

Keep audit trails and agent decision traces complete and usable for incident, risk and compliance review

Work with Security on agent identity, least privilege tool access, prompt injection response, data handling and secrets management

Monitor token and cost attribution by use case, and flag anomalies before they become a bill

Contribute to production readiness criteria, release planning and change management

Ensure compliance with enterprise standards and regulatory requirements

What you need to have

Below is an overview of the skills and experience we are looking for, but remember, don't rule yourself out if you don't have everything on the list. It's your intellect and your attitude we are really after.

Essential

4+ years of experience in application support, site reliability engineering or DevOps

Strong problem solving and debugging skills across distributed systems

Experience supporting large scale applications on AWS, including services such as Lambda, ECS/Fargate, API Gateway, Step Functions and CloudWatch

Experience with observability and monitoring tools such as Dynatrace or CloudWatch

Experienced in incident and problem management, including on-call support

Experience with CI/CD pipelines, Infrastructure as Code and automated deployments

Understanding of web, API and microservice technology

Working understanding of LLMs and AI agents, and how they fail differently from traditional software

Scripting skills in Python, Bash or similar

Excellent interpersonal and communication skills, including the ability to engage with stakeholders locally and in the US

Bachelor's degree, preferably in Computer Science, Engineering or a STEM subject. Comparable commercial experience may also be considered

Candidates must be based in Northern Ireland with the relevant Right to Work visa

Desirable

Experience supporting AI or LLM applications in production, particularly Amazon Bedrock

Familiarity with agent frameworks and protocols such as Bedrock AgentCore, Strands, LangGraph or MCP

Experience with AI evaluation, tracing or LLM observability tooling

Experience defining SLOs, error budgets and capacity planning

Cloud cost optimisation and FinOps experience, including token economics

Experience with event driven systems and messaging platforms (e.g. Kafka, AWS MSK)

Familiarity with security best practices and compliance standards in a regulated industry

Familiarity with ServiceNow, Jira, Confluence, Git and Azure DevOps

AWS certification or equivalent, including AI and machine learning certifications

So that's us. Thanks for taking the time to read this far. We look forward to hearing from you if you fancy running enterprise scale agentic AI in an innovative tech company that runs locally but is supported by a well established, highly recognised Fortune 500 company in the US.

Aflac Northern Ireland is