Full Stack AI Engineer

Accenture Southeast Asia; Singapore; On-site; SGD 180,000 - 240,000; Full time

Job summary

Accenture Southeast Asia seeks a Full Stack AI Engineer Manager to lead the technical delivery of agentic AI programs while staying hands-on. You will translate client requirements into agent designs and guide a team to deliver production-grade systems that create measurable business value.

As Forward Deployed Engineers, you will work directly with clients, bridge business problems to technical solutions, and evolve your team with rigorous code reviews, design sessions, and ongoing learning in

Qualifications

• Experience building production-grade, agentic AI systems.
• Ability to design end-to-end agentic solutions including harnesses and orchestration.
• Hands-on coding experience across full stack and AI tooling.

Responsibilities

• Architect and deliver agentic AI solutions end-to-end, contributing to code and design.
• Design agent harnesses, orchestration topologies, and LLM gateway configurations.
• Build and operate MCP-connected knowledge sources and RAG pipelines.
• Establish DevOps/AgentOps/LMMOps practices and governance.

Skills

LLM-based apps

Agent orchestration

Cloud-native

Python

Frontend skills

Education

Bachelor's degree

Master's degree

Tools

LangGraph

AutoGen

CrewAI

AWS Strands

Elasticsearch

Job description

The Full Stack AI Engineer leads the technical delivery of agentic AI programs while remaining an active, hands-on engineer. This role bridges the gap between business problems and technical solutions — working directly with clients to understand requirements, translating them into agentic application designs, and leading a team of engineers to deliver production-grade systems that generate measurable value.

Managers on this team operate as Forward Deployed Engineers — brought into client environments to rapidly understand a business problem, design a full stack AI solution, and build it. The expectation is genuine technical depth combined with client credibility: the Manager must be as effective in a code review or design session as they are in a client workshop. They develop their team, grow client relationships, and continuously evolve their knowledge of agentic AI as the field advances.

Position Responsibilities

Technical Solution Design and Hands-On Delivery

• Architect and deliver agentic AI solutions end-to-end — agents, orchestration, tool layers, knowledge pipelines, and full stack applications — at production engineering standards, contributing directly to code and design when required.
• Design agent harnesses, orchestration topologies (supervisor/worker, event-driven, parallel), A2A coordination patterns, and LLM gateway configuration across LLM providers.
• Build and maintain MCP servers, translate business processes into agent skills and reusable workflows, and implement advanced knowledge layer components: RAG, Text-to-SQL, Elasticsearch, knowledge graphs.
• Establish prompt architecture standards — versioning, A/B testing, structured output schemas — and apply reasoning patterns (ReAct, CoT, ToT) appropriate to each agent use case.

Agentic AI Technical Delivery

• Architect and build production agentic systems hands‑on — agent harnesses, orchestration topologies (supervisor/worker, event-driven, parallel), A2A coordination patterns, and LLM gateway configuration; write code, resolve complex engineering problems, and set the quality standard through personal example.
• Design and implement knowledge layer components: RAG pipelines (hybrid search, re‑ranking, late chunking), MCP‑connected knowledge sources, Text-to-SQL, Elasticsearch integration, and knowledge graph layers — selecting and tuning the right retrieval strategy per use case.
• Build and operate evaluation and AgentOps pipelines: golden datasets, LLM‑as‑judge, trajectory evaluation, agent testing suites (unit, integration, simulation), CI/CD for agents and prompts, asset registry management, production observability, and drift detection.
• Implement trust, safety, and governance components: guardrails, prompt injection defences, agent identity scoping, PII redaction, blast radius controls, HITL approval gates, and audit trail design for enterprise compliance.

Client Engagement and Business Translation

• Serve as the primary technical point of contact for client stakeholders — running workshops, translating business requirements into agent solution designs, and communicating technical decisions clearly to non‑technical audiences.
• Develop initial value hypotheses for agentic solutions — identifying automation and augmentation opportunities, estimating business impact, and establishing baseline metrics before delivery begins.
• Contribute to solution design and proposal development; identify expansion opportunities within current engagements and support account growth.

Delivery Excellence and AgentOps

• Lead workstream delivery in agile environments — managing scope, quality, technical risk, and milestone accountability with senior stakeholder visibility.
• Establish DevOps, AgentOps, and LLMOps practices: CI/CD for agent code and prompts, automated evaluation gates, deployment strategies, agent and asset registry management, and production operations.
• Define and implement evaluation frameworks, agent testing suites, HITL feedback capture, and production observability — distributed tracing, cost tracking, latency profiling, and drift detection.
• Implement guardrails, prompt injection defences, agent identity scoping, PII redaction, and audit trail design for enterprise compliance.

Team Leadership and Development

• Lead, manage, and develop a team of Consultants and Analysts — setting clear expectations, providing technical coaching, and running structured code reviews and design sessions.
• Foster a delivery culture of engineering rigour, continuous improvement, and learning — supporting team members in building agentic AI depth through challenging work and active knowledge sharing.

Innovation and Continuous Learning

• Maintain current, hands‑on knowledge of agentic AI developments — testing new frameworks, tooling, and research; bringing relevant advances into the team's engineering practice.
• Contribute to internal practice development: reusable accelerators, reference implementations, and delivery standards that improve capability across Accenture's AI practice.

Qualifications

• Building LLM-based applications in production — with operational accountability for deployed systems.
• Designing and delivering agentic AI systems — agents operating with meaningful autonomy in real production environments.
• Hands‑on experience with at least one agent orchestration framework (LangGraph, AutoGen, CrewAI, AWS Strands, or equivalent) in production.
• Demonstrated experience across the Agent Development Lifecycle: specification, harness build, tool and MCP integration, evaluation, deployment, observability, and refinement.
• Proven track record deploying software systems in production with measurable results — reliability, performance, or business value outcomes.
• Experience with knowledge layer engineering: RAG pipelines, MCP‑connected sources, Text-to-SQL, Elasticsearch, and knowledge graph design.
• AI/ML, data engineering, or advanced analytics — integrating intelligent systems into production software.
• full stack engineering: Python and a frontend framework (React, Angular, or Node.js); active, hands‑on capability across the stack.
• cloud‑native development on AWS, Azure, or GCP — CI/CD, containerised workloads, infrastructure as code, and production operations.
• Experience on complex digital transformation programmes — enterprise‑scale, multi-workstream, client‑facing delivery.
• Experience engaging directly with client stakeholders — translating business requirements to technical solutions.
• Bachelor's degree in a related field. A Master's degree is highly valued.