Agentic AI Engineer

We are seeking a hands-on Agentic AI Engineer to design, build, deploy, and operate secure AI-powered solutions that automate business processes, improve self-service capabilities, and accelerate decision-making. This role will focus on building enterprise-grade Agentic AI applications using AWS Bedrock AgentCore, Python, and the Strands framework while ensuring solutions are secure, observable, scalable, and production-ready. The engineer will be responsible for the full lifecycle of AI solutions, including development, deployment, evaluation, operations, continuous improvements, and production support. The position operates within a collaborative engineering environment and partners closely with API developers, product teams, operational teams, and business stakeholders.

Responsibilities:

• Design, develop, test, deploy, and support Agentic AI applications using AWS Bedrock AgentCore and Python.
• Develop agent orchestration workflows utilising approved tools, APIs, Model Context Protocol (MCP) servers, and enterprise integrations.
• Integrate AI agents with enterprise applications, knowledge repositories, workflow platforms, and cloud services.
• Implement security guardrails covering authentication, authorisation, tool permissions, prompt handling, data classification, logging, and compliance controls.
• Build automated evaluation frameworks that measure accuracy, safety, latency, cost efficiency, and task completion quality.
• Develop and maintain CI/CD pipelines using GitHub Actions and Harness.
• Create reusable frameworks, agent patterns, integration standards, prompt conventions, and reference implementations.
• Monitor production AI solutions and analyse failures, hallucinations, tool-call outcomes, performance trends, and operational risks.
• Maintain operational dashboards, runbooks, support documentation, and build- to-run processes.
• Collaborate with SMEs, engineering teams, and operational teams to ensure successful production adoption and ongoing support.

Requirements:

• Experience ( 5 to 9 years).
• 4+ years of software engineering, AI engineering, machine learning engineering, or related technology experience.
• Hands-on experience building and deploying AI or agent-based applications to AWS environments.
• Experience developing and supporting production software applications.
• Experience implementing Infrastructure-as-Code (IaC) for cloud deployments.
• Experience supporting applications after deployment in production environments.

Mandatory Technical Skills:

• AWS Bedrock and/or AWS Bedrock AgentCore
• Strands Framework
• Python
• Agentic AI application development
• API integration
• AI/LLM application deployment
• CI/CD implementation
• GitHub Actions
• Harness (or comparable enterprise CI/CD platform)
• Secure software engineering practices
• AI security and access-control concepts
• Production support and operational ownership

Required Project Experience:

Candidates must demonstrate experience with:

• Building and deploying agentic or LLM-based applications.
• AI agent orchestration and tool integration.
• Production deployment and support.
• Cloud-native deployments in AWS.
• CI/CD automation.
• Secure application development.

Good-to-Have:

• Retrieval-Augmented Generation (RAG)
• Vector databases/vector search
• Knowledge grounding solutions
• Agent evaluation frameworks
• Prompt-injection defence mechanisms
• AI red-teaming experience
• Human-in-the-loop workflows
• Kubernetes
• AWS cloud-security services
• CDK
• Helm Charts
• Cloud-native application architecture
• Cybersecurity solution development
• Experience within regulated industries
• Agile product delivery experience

Technical Skills:

Programming Languages:

• Python (Mandatory)
• Alternative Experience: Java, JavaScript, C++, AI / Agentic AI Technologies.
• Mandatory: AWS Bedrock, AWS Bedrock AgentCore, Strands Framework, Agentic AI Architecture, LLM Application Development
• Preferred: RAG, Agent Evaluation Frameworks, Knowledge Grounding, Vector Search, Human-in-the-Loop Controls, Prompt Defence Mechanisms, Cloud Platforms.
• Mandatory: AWS
• Preferred: AWS Cloud Security Services, DevOps / CI/CD
• Mandatory: GitHub Actions, Harness (or equivalent), CI/CD Automation, Infrastructure / Platform
• Mandatory Infrastructure as Code (IaC)
• Preferred: AWS CDK, Helm Charts, Kubernetes, Integration Technologies
• Mandatory: APIs, MCP (Model Context Protocol) or comparable integration methods, Security
• Mandatory: Secure Coding, Authorisation and Access Controls, AI Security Controls, Logging and Observability, Data Classification, Domain / Business Experience.

Preferred but not mandatory:

• Enterprise Technology
• Cybersecurity
• Cloud Engineering
• Financial Services
• Insurance
• Regulated Industries
• GCC / Global Delivery Environments
• Strong Agentic AI engineering experience is more important than industry background.

Behavioural / Soft Skills:

• Strong ownership mindset.
• Self-starter and proactive work style.
• Excellent communication skills.
• Strong collaboration across distributed teams.
• Ability to explain technical concepts to both technical and non-technical audiences.
• Problem-solving and analytical thinking.
• Strong prioritisation and time management.
• Accountability for production systems.
• Continuous learning mindset.
• Ability to work effectively with limited overlap across global teams.

Ideal Candidate Profile:

• A strong candidate is a hands-on AI Engineer with production experience building and deploying agentic AI applications in AWS environments.
• The individual should demonstrate expertise in Python, Bedrock AgentCore, Strands, AI integrations, and CI/CD automation while understanding security, governance, observability, and operational ownership.
• The ideal candidate is comfortable moving AI solutions from prototype to production and maintaining them throughout their lifecycle.