AI Engineer - Generative AI, LLM & RAG

1. Background

Xin Chao Science Edu is developing an AI Learning Intelligence Platform designed to transform public education through Generative AI, personalized learning, and data-driven educational intelligence.

The first phase of platform development has been completed and deployed on Amazon Web Services (AWS). The company is now entering the technical enhancement and validation phase, with a particular focus on integrating NVIDIA's Generative AI technologies, including Nemotron LLM, NVIDIA NIM, and NeMo Retriever for Retrieval-Augmented Generation (RAG).

The company is preparing a technical validation initiative in collaboration with NVIDIA Vietnam, Regional, and Global teams, targeting an initial Proof of Concept (PoC) across five public schools in Vietnam.

To strengthen its technical capabilities, Xin Chao Science Edu seeks an experienced Senior AI Engineer to lead the integration, optimization, and technical validation of its AI platform.

2. Overall Objective

The primary objective of this position is to enhance the technical performance, reliability, and intelligence of the existing AI Learning Intelligence Platform through the integration of NVIDIA Generative AI technologies and advanced RAG architecture.

The successful candidate will ensure that the platform can deliver accurate, curriculum-aligned, and personalized AI learning experiences while maintaining the technical reliability and scalability required for deployment in Vietnamese public schools.

3. Scope of Work and Key Responsibilities

A. Generative AI & NVIDIA Technology Integration

- Integrate NVIDIA Nemotron LLM and NVIDIA NIM into the existing AI platform.

- Implement and optimize NVIDIA NeMo technologies, including NeMo Retriever and relevant RAG components.

- Explore and implement NeMo Guardrails to improve AI response reliability and safety.

- Collaborate with NVIDIA technical teams in Vietnam, Regional, and Global offices.

- Evaluate and recommend appropriate LLM models and inference architectures.

B. RAG Architecture & Educational Content Integration

- Design and optimize RAG architecture for Vietnam's national curriculum and digital textbooks.

- Develop document processing pipelines, including PDF extraction, OCR, text cleaning, chunking, and embedding.

- Integrate and manage vector databases, including Qdrant.

- Improve semantic search, retrieval accuracy, and contextual response generation.

- Enable textbook-grounded AI tutoring, quiz generation, and learning material generation.

C. AI Learning Intelligence Development

- Enhance AI-powered student performance analysis.

- Develop personalized learning recommendations based on student learning data.

- Improve AI-generated student learning reports and learning insights.

- Optimize Vietnamese language understanding and curriculum alignment.

- Ensure the reliability and educational relevance of AI-generated responses.

D. AWS Infrastructure & AI Performance Optimization

- Work with the existing AWS-deployed architecture to integrate AI services.

- Optimize AI inference performance, response latency, and resource utilization.

- Implement monitoring and performance evaluation mechanisms.

- Support secure and scalable AI service deployment.

- Coordinate with existing backend and platform developers for system integration.

E. Technical Validation & PoC Implementation

- Prepare technical architecture and validation plans for the five-school PoC.

- Conduct LLM and RAG performance testing, including accuracy, retrieval quality, and response reliability.

- Evaluate system performance under concurrent user scenarios.

- Identify technical risks and implement corrective measures.

- Prepare technical validation reports and recommendations for commercialization and scaling.

4. Expected Deliverables

Key technical outputs:

1. NVIDIA LLM integration into the existing platform.

2. Functional RAG pipeline using Vietnamese curriculum and textbook content.

3. Improved AI Tutor and textbook-based learning content generation.

4. AI response accuracy and performance evaluation framework.

5. Technical documentation and PoC validation report.

5. Required Qualifications and Experience

Educational Background

- Bachelor's or Master's degree in Computer Science, Artificial Intelligence, Software Engineering, or a related field.

Professional Experience

- Minimum 3–5 years of relevant software engineering or AI development experience.

- Demonstrated experience in Generative AI, LLM application development, and RAG implementation.

- Hands-on experience deploying AI applications in production environments.

- Experience integrating AI models with existing software platforms.

Technical Skills

Technical Area | Required Competency

Programming: Python

Generative AI: LLM, Prompt Engineering

RAG: RAG Architecture, Retrieval Optimization

Vector Database: Qdrant, Milvus, or Pinecone

NVIDIA AI: NIM, Nemotron, NeMo (preferred)

Cloud: AWS

Backend: FastAPI, REST APIs

Deployment: Docker, CI/CD

AI Evaluation: RAG Evaluation, LLM Performance Testing

6. Preferred Qualifications

- Experience with NVIDIA AI technologies and NVIDIA developer ecosystems.

- Experience with educational technology, EdTech, or AI-powered learning platforms.

- Experience processing digital textbooks, OCR, and multilingual educational content.

- Experience with Vietnamese language NLP or Southeast Asian language models.

- Ability to communicate technical concepts effectively in English.

- Experience collaborating with international technical teams.

7. Key Performance Indicators (KPIs)

The engineer's performance will be evaluated based on the following indicators:

1. NVIDIA Integration: Successful integration and functional validation of NVIDIA LLM technologies.

2. RAG Performance: Improved retrieval accuracy and grounded AI responses using Vietnamese textbook content.

3. AI Functionality: Successful implementation of textbook-based AI tutoring and learning content generation.

4. System Reliability: Stable AI services with measurable response latency and system performance.

5. PoC Readiness: Completion of technical validation and readiness for deployment across five public schools.

6. Technical Documentation: Delivery of architecture documentation, testing results, and technical reports.

Specific quantitative targets will be agreed upon during the first month following the technical baseline assessment.

8. Working Arrangements & Collaboration

The Senior AI Engineer will work closely with:

- CEO / Founder – Strategic direction and product priorities.

- NVIDIA Vietnam Technical Team – Technical architecture and implementation support.

- NVIDIA Regional and Global Teams – Technical consultation and validation.

- AWS Team

The engineer is expected to take ownership of assigned technical deliverables and proactively communicate technical challenges, risks, and solutions.

9. Expected Outcomes

By the end of the initial three-month assignment, Xin Chao Science Edu expects to achieve:

- A technically enhanced AI Learning Intelligence Platform.

- Successful integration of NVIDIA Generative AI technologies.

- A functional RAG-based educational knowledge system using Vietnamese curriculum content.

- Improved AI tutoring and personalized learning capabilities.

- A technically validated platform ready for the five-school public education PoC.

- A scalable technical foundation for nationwide commercialization and future regional expansion.

10. Future Growth Opportunity

The successful candidate will have the opportunity to grow into an AI Technical Lead role, taking increasing responsibility for technical architecture, AI product development, engineering strategy, and future technical team leadership as the company expands. Hello Science Edu is the member of NVIDIA Inception and currently accelerating programme of Plug and Play.