AI Engineer at Global Sourcing… at OneTeamAnywhere · Singapore

We are looking for an AI Engineer to work with us in a remote setup for our Singapore-based client.

About the OTA Client

An AI sourcing agent for consumer brands. Describe what you want to launch — a hydrating lip balm, custom gym apparel, frosted glass dropper bottles — and the agent comes back with verified suppliers, real quotes, and DDP pricing in under 24 hours. What used to take weeks of Alibaba browsing now takes one chat.

Founded in 2021. Today they serve 200+ brands with a network of 4,000+ verified suppliers across 12 countries and a 98% on-time delivery rate. They're headquartered in Singapore with entities in the US, Hong Kong, and Indonesia, and they're now pushing hard into LatAm and the US.

The role

A multi-agent system in production. The agent talks to suppliers, negotiates pricing, tracks specs across dozens of factories, and keeps brands updated in chat. There's a long list of things it does well — and a longer list of things it should do next.

You'll own meaningful surface area of that stack. You'll read papers on Friday, ship the primitive on Monday, and watch it run against real supplier conversations on Tuesday.

What you'll own

• Agentic development — building, hardening, and extending a multi-agent sourcing system

• Observability and evaluation — the infra that tells us when the agent is doing the right thing, and when it isn't

• Tools and integrations — wiring the agent into supplier systems, document workflows, and our own ops

• Experimentation — new models, memory architectures, context strategies — proposed and shipped by you

What we're looking for

• Deep, hands-on work on agentic / multi-agent systems — you've built and shipped them, not just read about them

• AI-native, AI-first — your default coding workflow is AI-assisted. You can articulate why and where it works.

• Constantly learning — you experiment with new AI concepts proactively. You don't wait to be told what to try next.

• Experience building observability and evaluation systems for AI products — traces, evals, replay, regression detection

• Practical builder mindset — we'd take a strong builder over a pure research profile every time

Nice to have

• Production experience with LLM memory, persistent context, or context engineering at scale

• Open source contributions in the agent / eval space

• Background working across time zones — a lot of our supplier and brand traffic is global