Platform Engineer – AI Platform

About the role

We are looking for a Platform Engineer to help build and operate an AI platform that serves employees across the organization. You will work in a hybrid setting in Bangkok, collaborating with cross‑functional teams to deliver reliable, secure AI services.

Key responsibilities

• Develop and maintain AI agents, onboard new users, add integrations, and iterate based on feedback.
• Enhance the AI Gateway by adding data sources, writing OPA policies, expanding retrievers, and keeping audit logs.
• Improve the RAG pipeline: ingest documents, boost embedding quality, label data sensitivity, and measure answer accuracy.
• Manage LLM serving on GPU servers, evaluate new models, handle context length, memory budgets, and integrate external APIs responsibly.
• Design and deliver automated AI workflows (e.g., report summarisation, support assistance, data analysis) from proof‑of‑concept to production.
• Own end‑to‑end delivery: design, build, test, deploy, and support post‑launch, tracking usage metrics.
• Write documentation and train employees on safe, effective AI tool usage.

Required profile

• Minimum 2 years of professional programming experience, strong Python skills, and comfort with Linux/command‑line environments.
• Fluent English, both spoken and written.
• Hands‑on experience building real solutions with large language models (LLMs), including API calls, RAG, agents, or fine‑tuning.
• Deep understanding of LLM fundamentals such as tokens, context windows, embeddings, prompt engineering, and hallucination mitigation.
• Knowledge of AI security concerns like prompt injection and data leakage, and the need for an authorization layer.
• Fast learner who can work with unfamiliar systems and read existing code/documentation.

Required skills

• Python
• Linux command line
• Open Policy Agent (OPA)
• vLLM / Ollama
• GPU computing
• API integration
• RAG pipelines and embeddings
• Prompt engineering
• AI coding assistants (Claude Code, Codex, Copilot)

What we offer

• Hybrid work model
• Competitive remuneration based on experience
• Hands‑on work with real GPU servers, models, and users
• Clear growth path toward lead engineer of the AI platform