
Data and AI engineers
Already building with models and ready to move from prompts to agents that act.
Three days building AI agents that take action, from a single agent to a team of them running real workflows — at your office or online, for your team only.
Every day ends with something running. Nothing is pre-recorded.
Agent architecture and the OpenAI SDK, then a working agent of your own: memory, tokens, traces and tool calls. By the end of the day your team can see exactly what an agent did and why.
Multi-agent design — roles, handoffs, guardrails and tracing across the group — with AutoGen and LangGraph introduced side by side so the trade-offs are clear.
CrewAI and MCP, workflow automation with n8n, and integrations into external systems. The three days close with a hands-on project built against one of your own business use cases.
This one is technical — it is built for the people who will own the agents afterwards.

Already building with models and ready to move from prompts to agents that act.

Own the workflows that agents will plug into, and the failure modes when they do.

Building the product themselves and need agents working without hiring a team first.
Tell us your team and your goals. We will recommend the right shape and, for Malaysian employers, prepare the HRDC submission.
Thanks for applying — we read every one, and we’ll be in touch about next steps.
AMASTSoftware that runs your operations — built and implemented by a team that stays with you.
San Francisco · London · Dubai · Kuala Lumpur

