Job Description

About the client

Our client specialises in building Agentic AI systems.

Our client is looking to onboard a Platform Engineer to serve as one of the principal architects of SupplyOS, their agentic AI platform that powers procurement, inventory, manufacturing, and logistics operations across Asia.

You will help define the long term technical direction of a system that enables enterprises to run faster, smarter, and more automated workflows. You will shape their core architecture, build the platform primitives used by all teams,

What you'll be doing

• Design and build major platform capabilities like our API server, workflow engine, multi-agent orchestration, RAG infrastructure, and intelligent memory systems that form the backbone of SupplyOS.
• Constantly identify recurring patterns across deployments and crystallise them into platform abstractions that make FDEs radically faster.
• Establish engineering standards for reliability, testing, observability, performance, and developer experience.
• Evaluate and integrate cutting-edge AI technologies such as new agent frameworks, LLM providers, vector databases and determine what deserves a place in their stack.
• Solve the hard problems of reliability, performance, and cost optimization as they grow from handling thousands to millions of supply chain events.
• Partner with FDEs and product teams to translate field learnings into platform capabilities, ensuring their abstractions match real-world complexity.

Requirements

• Experience: 5+ years
• You've architected backend systems that handle real scale and complexity and you understand consensus protocols, eventual consistency, and the subtle tradeoffs that separate systems that work from systems that work reliably and deliver enterprise workloads.
• You ship production-grade software in Go, Python, and JavaScript/TypeScript. You have gone beyond "can write code" to "writes code that survives production at 3am".
• You're fluent in building production cloud systems with Terraform and Docker. Infrastructure-as-code isn't just a buzzword, it's how you think about deploying and scaling systems.
• You have a proven track record of building platforms that multiple teams actually use and depend on and you've felt the pain of bad abstractions and know what makes developers productive.
• You understand how to move, transform, and validate data at scale. ETL isn't a mystery, it's a solved problem you've implemented many times. Bonus points if you've wrestled with Databricks in production