Building Production-Ready AI Agents
Your agent works in the demo. But what happens when real users start hitting it?
Join us for a hands-on Tinker Time with Bitdeer AI to build and deploy an agent directly on their cloud and GPU infrastructure.
Then, place your under load and see what starts to break. Work through the infrastructure and engineering decisions behind running agents reliably, from compute and deployment to concurrency, cost, observability, and evaluation.
🛠️ What We'll Work Through
Build & Deploy
Structure your agent, provision the right compute, and take it from your local environment to a live, callable endpoint.
Scale & Stress-Test
See what happens under parallel requests and work through concurrency, latency, failure modes, and the trade-offs between performance and inference cost.
Observe & Evaluate
Add logging, tracing, retries, and repeatable evals so you can understand how your agent behaves rather than assume it's working.
Beyond the Workshop
Think through what still needs to happen before an internal pilot or production release, including security, data handling, reliability, and operational considerations.
⚠️Some Important Notes
Please bring an internet-facing laptop for the session.
As this is a technical workshiop, working familiarity with Python and APIs will be required.
More About the Facilitator: Bitdeer AI
Bitdeer AI provides cloud and GPU infrastructure for training, deploying and scaling AI workloads across Asia Pacific, giving teams access to high-performance compute without the overhead of building and maintaining their own environments.
More About the Series
Tinker Time is Lorong AI's hands-on meetup combining guided workshops with open collaboration time for builders to share code, troubleshoot together, and experiment with AI development techniques.
Get involved: Learn more about Lorong AI | Speaker Sign-up | WhatsApp Community | LinkedIn | X