AI Engineer

You will build AI capabilities embedded directly inside an ERP platform used by thousands of businesses.

Instead of working on isolated demos, you will help deliver production AI features used in daily operations.

– Develop AI-powered features such as: AI Chat Assistant, Document AI, Data Analysis AI, Workflow Automation Agents

– Design and implement agent workflows using LangGraph: task planning agents, tool-calling agents, multi-step reasoning workflows

– Work with multiple LLM providers including: OpenAI, Gemini, Qwen, Ollama

– You will help implement: model routing, prompt templates, fallback strategies

– Develop pipelines that turn enterprise data into AI-ready knowledge: text extraction, embedding generation, indexing, semantic search

– Core data systems: MongoDB, Elasticsearch

– Implement background processing for AI workloads using: Celery workers, Kafka/RabbitMQ messaging

– Used for: document processing, AI tasks, agent execution

– Use observability tools to monitor AI performance. Tools include: Langfuse, OpenTelemetry

– You will work on: tracing LLM calls, debugging prompts, performance analysis

– Competitive and negotiable compensation, aligned with your experience and capabilities.

– End-to-end product involvement – participate in the full product development lifecycle and grow alongside the continuous expansion of the 1Office platform.

– Exposure to diverse clients, from fast-growing SMEs to large enterprise groups, along with engaging monthly and quarterly networking and team activities.

– Collaborate with a passionate and highly skilled Product Builder team that has successfully developed the 1Office platform, currently serving 6,000+ enterprise customers and over 500,000 monthly users.

– Full statutory benefits in accordance with Vietnamese Labor Law, including Social Insurance (SI), Health Insurance (HI), and Unemployment Insurance (UI).

– Holiday bonuses on major occasions such as Reunification Day (April 30), International Workers’ Day (May 1), National Day (September 2), New Year, and Lunar New Year (Tet).

– Professional development opportunities, including technical and soft-skill training programs, as well as quarterly company activities.

– Opportunity to be trained in automation testing, supporting long-term career development.

– Modern and premium office environment, featuring comfortable working spaces and recreational facilities such as a billiards table for breaks.

– Annual company trips and team-building activities, creating memorable experiences with colleagues.

– Many additional attractive perks and benefits designed to support employee well-being and engagement.

Experience

– 2+ years of experience in Software Engineering

– Experience working with LLM APIs (such as OpenAI, Gemini, or similar models)

– Experience building backend services or production APIs

Core Skills: You should have experience working with the following technologies

– Python as the main programming language for backend and AI services

– REST APIs and backend service development

– LLM API integration (OpenAI, Gemini, or similar)

– MongoDB for application data storage

– Elasticsearch for search and analytics

– Prompt engineering to optimize LLM outputs

AI Knowledge: Good understanding of common concepts used in modern AI systems: RAG (Retrieval Augmented Generation), Embeddings and vector representations, Semantic search, AI agents and tool-calling workflows

– These are not required but will be a strong advantage: LangGraph for building agent workflows, Langfuse for LLM observability and prompt analytics, Kafka or RabbitMQ for messaging systems, Celery for asynchronous task processing, OpenTelemetry for distributed tracing and monitoring.

Tech Stack: You will work with a production AI stack designed for a SaaS platform.

Backend: Python, FastAPI

AI Stack: LangGraph – agent orchestration, Langfuse – LLM tracing and observability, Mem0 – AI memory system, MCP (Model Context Protocol) – AI interaction layer

LLM Providers

– The system supports a multi-model architecture, including: OpenAI, Gemini, Qwen, Ollama (local models)

– Data Platform: MongoDB – primary application database, Elasticsearch – search and analytics

– Messaging & Async Processing, AI workloads are handled asynchronously using: Kafka, RabbitMQ, Celery

– Observability: OpenTelemetry for distributed tracing and monitoring

– Infrastructure: Docker, Linux

• Email: hr@1office.vn
• Hotline: 1900 3313