Full Stack AI Developer

Accenture Southeast Asia; Singapore; On-site; SGD 90,000 - 130,000; Full time

Job summary

Accenture is seeking a Full Stack AI Consultant (Developer) in Singapore to build production-grade agentic AI applications. You will design backends with Python, create frontend interfaces, and deploy reliable MCP servers and knowledge pipelines.

You will implement RAG, memory architectures, and drive end-to-end DevOps practices within client delivery teams. You will work with LangGraph, AutoGen, and CrewAI to craft scalable agent ecosystems, while translating business needs into technical

Qualifications

• 3+ years of full stack software engineering in production.
• 3+ years building and deploying LLM-based applications.
• Hands-on experience with agentic AI frameworks and tool integration.
• 2+ years cloud-native development on AWS, Azure, or GCP.
• Bachelor's degree in Computer Science, Engineering, or related field; Master’s valued.

Responsibilities

• Design and build full stack agentic AI applications to production standards.
• Develop MCP servers, tool integrations, and knowledge pipelines.
• Implement RAG pipelines, memory architectures, and efficient prompts.
• Maintain CI/CD, observability, and automated testing for AI systems.
• Collaborate with managers/clients to turn requirements into technical specs.

Skills

Full stack development

LLM applications

Agentic AI frameworks

AI/ML integration

CI/CD

Cloud-native development

Education

Bachelor's degree in Computer Science

Master's degree valued

Tools

LangGraph

AutoGen

CrewAI

MCP servers

Elasticsearch

Job description

About Accenture Data & AI

The beginning of a new Data & AI decade that will reshape work and society is underway. Accenture is stepping boldly into this future with a clear strategy and purpose: to help clients optimize and reinvent their businesses with data and AI — backed by a $3 billion investment and a commitment to industry-defining work.

About Accenture Data & AI

The beginning of a new Data & AI decade that will reshape work and society is underway. Accenture is stepping boldly into this future with a clear strategy and purpose: to help clients optimize and reinvent their businesses with data and AI — backed by a $3 billion investment and a commitment to industry-defining work.

With over 45,000 professionals dedicated to Data & AI, Accenture's Data & AI organization brings together Experienced Innovation, Strategic Investment, Exceptional Talent, and a Power Ecosystem to deliver outcomes at the frontier of what is possible.

About The Role

The Full Stack AI Consultant (Developer) is a hands‑on engineer at the core of Accenture’s agentic AI delivery capability. This role is for those who build things — full stack applications, agentic workflows, knowledge pipelines, and tool integrations — and who want to do that work at the frontier of enterprise AI. Consultants work within delivery teams, turning business requirements into production‑grade agentic AI solutions under the guidance of technical leads and managers.

The expectation is active, daily engineering: writing code, building agents, implementing MCP servers, designing knowledge pipelines, and maintaining the DevOps and AgentOps practices that keep production systems running. Consultants on this team are also expected to invest continuously in learning — agentic AI is evolving rapidly, and staying current is a professional responsibility, not optional.

Position Responsibilities

Full Stack Application Development

• Design and build full stack agentic AI applications — Python backends, REST and event‑driven APIs, and React or equivalent frontends — to production engineering standards.
• Implement agentic application UX: streaming responses, intermediate output display, reasoning transparency, and error and escalation interfaces for end users.
• Translate business requirements into technical specifications; work with managers and clients to clarify scope, surface ambiguities, and deliver against agreed outcomes.

Agentic AI Development

• Build and configure agents using established orchestration frameworks (LangGraph, AutoGen, or equivalent): harness setup, persona and instruction loading, tool binding, memory configuration, and lifecycle management.
• Implement reasoning patterns (ReAct, Chain‑of‑Thought, Plan‑and‑Execute) appropriate to each agent use case; design and version prompt architecture including system prompts, few‑shot examples, and structured output schemas.
• Build multi‑agent workflows — defining agent roles, A2A handoff contracts, shared state schemas, and escalation paths — under architectural guidance from technical leads.

MCP, Tools, Skills, and Workflows

• Design, build, and maintain MCP servers connecting agents to enterprise systems, APIs, databases, and SaaS platforms — with robust schema design, error handling, idempotency, and retry logic.
• Translate business processes into agent‑executable skills, structured instructions, and reusable workflows — bridging the gap between business requirements and agent implementation.
• Implement context engineering pipelines, memory architectures (episodic, working, long‑term), and LLM gateway configuration to support reliable, cost‑efficient agent operation.

Knowledge Layer Implementation

• Build RAG pipelines: document ingestion, chunking, embedding, vector store indexing, hybrid retrieval, re‑ranking, and quality evaluation.
• Implement Text‑to‑SQL capabilities — schema grounding, query generation, validation, and safe execution against enterprise databases.
• Integrate Elasticsearch as a retrieval backend; build knowledge graph components and ontology‑driven query layers where required by the use case.

DevOps, AgentOps, and Quality

• Maintain CI/CD pipelines for agent code, prompt changes, and infrastructure — including automated evaluation gates and deployment strategies across environments.
• Instrument agentic systems with production observability: distributed tracing, token cost tracking, latency profiling, failure logging, and drift detection.
• Build agent testing suites: unit tests with mocked tools, multi‑agent integration tests, and simulation environments; implement guardrails, PII redaction, and audit trail logging.
• Manage versioned agent and asset registries — agents, tools, skills, prompts, and workflows — with controlled promotion across development, staging, and production.

Continuous Learning and Innovation

• Maintain active, current knowledge of agentic AI frameworks, tooling, and research — testing new approaches and bringing relevant innovations into the team's engineering practice.
• Contribute to internal knowledge sharing: documenting patterns, building reusable accelerators, and supporting capability development within the team.

Core Requirements

• 3+ years of full stack software engineering in production — Python backend and a frontend framework (React, Angular, or Node.js).
• 3+ years building and deploying LLM‑based applications — prompt engineering, RAG, tool integration, or agent development.
• Hands‑on experience with agentic AI frameworks — agents, tools, orchestration — in project or production environments.
• 3+ years experience in classical AI/ML, data engineering, or analytics — integrating models or data pipelines into software applications.
• Experience with classical software engineering disciplines: CI/CD, TDD, version control, code review, and automated testing.
• 2+ years cloud‑native development on AWS, Azure, or GCP — containerised workloads, managed services, and infrastructure as code.
• Experience translating business requirements into technical specifications and delivering against them.
• Bachelor's degree in Computer Science, Engineering, or a related field. A Master's degree is highly valued.

Additional Skills

• Hands‑on experience building agents with LangGraph, AutoGen, CrewAI, or equivalent frameworks.
• Experience building MCP servers or tool integration layers for LLM applications.
• Experience implementing RAG pipelines — ingestion, retrieval, re‑ranking, and quality evaluation.
• Exposure to Text‑to‑SQL, knowledge graphs, or Elasticsearch for enterprise knowledge retrieval.
• Experience with DevOps/AgentOps/LLMOps tooling and CI/CD for AI systems.
• Experience with classical AI/ML model integration — prediction APIs, model serving, feature pipelines.
• Active engagement with the agentic AI community — open source contributions, side projects, or published learning.
• Proficiency with AI‑assisted development tools (Claude Code, GitHub Copilot, Codex).