AI / Agent Engineer

About MixWork & Our Client

MixWork is the premier HR outsourcing partner for strategic workforce solutions, empowering brands and global organizations with skilled, dedicated, and professional top-tier regional teams from South East Asia to accelerate their business growth. On behalf of our client, a premier, household-name retail enterprise headquartered in Singapore with a well-established international footprint and a heritage of quality and operational excellence, we are seeking a dedicated professional to join their team as they actively modernize their omni-channel capabilities and leverage a sophisticated regional technology footprint to expand their enterprise AI ecosystem, building intelligent agent solutions, scalable AI platforms, and integrated data intelligence capabilities across their retail operations.

Role Summary

Own the client’s AI and agentic platform, built on Amazon Bedrock + Claude, as the primary platform. The client operates a hybrid AI architecture: AWS Bedrock handles the data intelligence layer (conversational agents, analytics, MCP servers, Claude Cowork integration, customer-facing AI), while Azure AI capabilities embedded in D365 and Power Platform (Copilot Studio, AI Builder, D365 Copilot) handle enterprise process automation and are delivered by client’s Microsoft implementation partner. The AI/Agent Engineer owns the AWS platform fully and is responsible for integrating outputs and data flows between the AWS and Azure AI layers.

Key Responsibilities

• Own and operate all Amazon Bedrock agents: the Operations Agent (store floor product and brand knowledge), the Org-Wide Q&A Chatbot (live data queries across both MCPs), and the Scheduled Email Reporting Agent (automated daily/weekly analytics briefs via Amazon SES).

• Own and maintain both MCP servers: the Data Query MCP (natural language → SQL → Athena Gold views, read-only, governed access) and the Knowledge Base MCP (RAG retrieval from Bedrock Knowledge Base over ops documents and Gold layer data).

• Maintain the Bedrock Knowledge Base: Aurora Serverless v2 + pgvector, knowledge sync cadence, S3 ops document governance, and retrieval quality.

• Own the Claude Cowork MCP server: OAuth endpoint exposing client’s Gold layer to Claude Cowork for management use; maintain token lifecycle and IP allowlist.

• Build and deploy post-ECI agents: Chat Web App (React/Next.js, Cognito SSO), Scheduled Email Reporting Agent, and Org-Wide Q&A Chatbot.

• Develop complex multi-agent orchestration: agents that can execute write-actions across client’s systems (D365 triggers, e-commerce updates, marketing automation) with appropriate approval gates.

• Build customer-facing agents: AI shopping assistant and loyalty concierge on the e-commerce storefront, in collaboration with the Shopify Plus Developer.

• Integrate Azure AI outputs into the AWS platform: design and implement the data and event flows that connect Azure AI outputs to the AWS Gold layer, CDP, and Bedrock agents.

• Maintain Terraform infrastructure-as-code for the AI platform; manage ECS Fargate services; optimise Bedrock token costs and AgentCore Runtime.

• Shadow CloudMile during ECI build; produce client’s internal AI platform runbooks before handover.

Required Qualifications & Experience

• Minimum 2-3 years of software engineering experience, including at least 2 years of dedicated ML/AI experience (specifically building RAG/company brain systems).

• Demonstrable hands-on production experience with Amazon Bedrock (Agents, Knowledge Base, Guardrails) as the primary requirement.

• Strong Python: Lambda development, API client development, prompt engineering, structured LLM output parsing.

• Experience building and deploying containerised applications (Docker, ECS Fargate or equivalent).

• English proficiency at B2 minimum (IELTS 6.0); mandatory assessment at interview.

• Demonstrated independence on complex, open-ended technical problems: able to debug agent reasoning failures, design agent topologies from requirements, and deliver working systems without architectural hand-holding.

Technical Proficiency

• Amazon Bedrock: Agents, Knowledge Base (Aurora pgvector), Guardrails, cross-region inference profiles, AgentCore Runtime.

• MCP (Model Context Protocol): building MCP servers as Lambda + Function URL; tool definition; OAuth authentication.

• AWS: Lambda, EventBridge, SES, ECS Fargate, ALB, CloudFront, S3, Amazon Cognito, WAF.

• React or Next.js: sufficient to own and extend the Chat Web App and contribute to the Mobile PWA.

• Amazon Athena: able to write and validate SQL for MCP query generation.

• Terraform: read, modify, and apply infrastructure-as-code for the AI platform.

• Azure AI awareness (working knowledge, not build-and-own): Microsoft Copilot Studio, Azure OpenAI Service, Power Automate AI Builder, and D365 Copilot.

Nice to Have

• AWS Certified Machine Learning – Specialty.

• Bedrock Intelligent Prompt Routing.

• Agentic frameworks: LangGraph, AutoGen, or CrewAI.

• Customer-facing AI integration on e-commerce platforms.

• Hands-on experience with Microsoft Copilot Studio or Azure AI Foundry.

• Experience with third-party vendor coordination and strong English communication skills.

Statutory Provisions & Compliance

• Full Legal Compliance: Official employment contract managed under MixWork Indonesia, ensuring complete adherence to local labor laws and employment standards.

• Healthcare & Social Security: Full registration and contributions for both BPJS Kesehatan and BPJS Ketenagakerjaan to ensure comprehensive coverage.

• Religious Holiday Allowance: Guaranteed annual mandatory Religious Holiday Allowance (THR) paid in accordance with statutory government regulations.

What We Offer

• Flexible Medical Benefit: Comprehensive healthcare coverage fully inclusive of dental, optical, outpatient care, and wellness treatments to support your overall well-being.

• Daily Allowances: Competitive transportation and meal allowances to support your operational needs.

• Workstation Provisioning: High-performance corporate laptop and necessary technical equipment provided.

• Regional Ecosystem: Access to ongoing global corporate alignment, dedicated HR support, and a stable, creative career trajectory with a premier international brand.

Important Notes

• Language Requirement: As this is an international role, please note that all screenings and interviews will be conducted exclusively in English.

• Equal Opportunity Employer: MixWork is committed to creating an inclusive, diverse, and fair workplace culture. We value talent and capability above all else, completely free from discrimination or bias.