#EG Fullstack Developer – AI Applications

This role sits within NCS AI Central's (AIC) Forward Deployed Engineering (FDE) model — the combined capability that takes AI solutions from proof-of-concept through to hardened production systems. As Fullstack Developer, you build both the frontend and backend of Gen AI applications — from UI through API and cloud deployment — operating across both fast-moving FDE engagements (POC/POV, pilot deployments) and steady-state system development and maintenance work.
What will you do:
1. Frontend Development
Build a simple, clean UI to support key POC and application workflows (upload, query, view results), using ReactJS.
Implement screens with usable UX, optimised for performance and cross-device compatibility.
Integrate with backend APIs for model inference, RAG results, and agentic workflow status and outputs.
Handle user inputs, file uploads, and display model/agent outputs responsively — accounting for the multi-step, non-deterministic nature of agentic workflows.
Add lightweight feedback/error handling for API failures or delayed responses; implement basic authentication or access control where required.
2. Backend Development
Design and develop backend services and APIs using Python (FastAPI/Flask) to support Gen AI functionality.
Implement endpoints for model inference requests, RAG workflows, agentic workflow orchestration (e.g. using frameworks such as AWS Strands or LangGraph), and document upload/processing/storage.
Integrate with cloud AI services for inference and model orchestration — AWS Bedrock, Azure AI Foundry, or Google Vertex AI depending on the engagement — including China-origin models (DeepSeek, Qwen, GLM) where self-hosted or exposed via compatible endpoints. Build lightweight pipelines for RAG (chunking, embeddings, retrieval).
Apply LLMOps practices to support model monitoring and versioning as applications mature toward production.
Manage data storage using Postgres (RDS or equivalent) and/or vector search services (OpenSearch, FAISS, or cloud-native equivalents).
3. Deployment & Quality
Containerise applications using Docker and deploy to AWS ECS Fargate/Lambda, Azure Container Apps/Functions, GCP Cloud Run, or equivalent Kubernetes-based services (EKS/AKS/GKE) where required.
Implement CI/CD pipelines for rapid iteration, using infrastructure-as-code (Terraform or CDK) where practical.
Perform code scans and reviews to identify security issues early (e.g. SonarQube, Fortify).
4. FDE & Development/Maintenance Coverage
During FDE engagements: prioritise speed, clarity, and flexibility over perfect design, iterating quickly based on user feedback and validation sessions.
During system development & maintenance engagements: harden the same application into production-grade code, with full CI/CD, monitoring, and security scanning in place.
5. Collaboration
Work closely with the Application Architect and Cloud Engineer/Architect to align API contracts and architecture.
Document API contracts, workflows, and deployment/UI steps for the team.
Role Levels We Are Hiring For
We are hiring at two levels for this role. All responsibilities above apply to both; the distinction is in scope of ownership, years of experience, and seniority of judgement expected.
Fullstack Developer – AI Applications
2–4 years of relevant experience. Builds frontend and backend features for one engagement at a time, under guidance from a Senior Fullstack Developer or Application Architect.
Executes defined API/UI work; escalates architecture-level design decisions to the Application Architect.
Senior Fullstack Developer – AI Applications
5+ years of relevant experience, including at least 2 years hands-on in AI/Gen AI application development. Owns fullstack delivery across more complex engagements and works more independently with the Application Architect on design decisions.
Mentors junior Fullstack Developers and sets implementation patterns others follow.
The ideal candidate should possess:
Strong in ReactJS, basic state management (Context, Redux, or simple React Query), and TypeScript.
Strong Python development experience (FastAPI/Flask).
Experience with REST APIs, JSON, and async handling.
Hands-on experience with at least one major cloud AI service (AWS Bedrock, Azure AI Foundry, or Google Vertex AI), including model integration, orchestration, and prompt handling.
Working knowledge of China-origin models (DeepSeek, Qwen, GLM) a plus.
Familiarity with agentic orchestration frameworks such as AWS Strands or LangGraph, for implementing multi-step agent workflows.
Familiar with file upload handling (PDF, DOCX, text).
Containerization experience (Docker) and deployment to at least one major cloud's container/serverless services; familiarity with Kubernetes (EKS/AKS/GKE) is a plus.
Familiarity with Postgres schema design, querying, and managed database setup.
Comfortable building lightweight RAG and agentic pipelines and integrating with vector stores.
Understand how Gen AI and agentic outputs can vary and need dynamic rendering.
Working knowledge of secure coding practices and code scanning tools.
Preferred Qualifications
Experience with OpenSearch Serverless, Azure AI Search, or other vector databases.
Familiarity with event-driven architectures (SQS/SNS/EventBridge or Azure/GCP equivalents).
Basic knowledge of prompt engineering and Gen AI output evaluation.
Exposure to infrastructure-as-code tools (Terraform).
Exposure to Singapore Government GCC/HCC environments.
Tech Stack (Illustrative)
Languages: Python, TypeScript/JavaScript
Frontend: ReactJS, Redux
Backend: FastAPI/Flask
Platform: Docker, AWS ECS Fargate/Lambda, Azure Container Apps/Functions, GCP Cloud Run, EKS/AKS/GKE (as needed)
LLM Runtime: AWS Bedrock, Azure AI Foundry, Google Vertex AI, AWS Strands/LangGraph; DeepSeek/Qwen/GLM (China stack)
Storage/Search: Postgres (RDS or equivalent), OpenSearch, FAISS
CI/CD: Terraform/CDK, standard pipelines
Code Quality: SonarQube, Fortify

Employment type
Full-time

Work arrangement
No