Agentic - ai@govtech

Enter GovWork-Bench, our take on what an eval for Public Service work could look like.

The Lab

GovWork-Bench – A Singapore Public Service Benchmark to evaluate AI Assistants on Public Service tasks

Evals are all the rage in the AI space recently, especially with news of GPT-6 Astra saturating long-standing difficult benchmarks like ARC-AGI-3 and FrontierMath (Tier 4). In this article, we too are sharing our latest work on benchmarking — but with a unique Public Sector twist!

AgenticEvals

The Lab

Agent Wars, Episode 3: Revenge of the Traces

On observability as the layer agentic AI deployments may miss, and it's importance in deciding whether you can trust anything else you build

Agentic

The Lab

Strengthening Chatbot Safeguards with Red-Teaming Agents

How we used agentic AI to proactively test and strengthen chatbot safeguards for chat.gov.sg

AgenticSecurity

The Lab

When Agent Skills Evolve: The Need for a Layer of Control

What we learnt from building Rhizomatic, a human-in-the-loop proof of concept for reviewing, versioning and rolling back changes to reusable agent skills.

AgenticInfrastructure

The Studio

ORCA Part 2: Under the Hood

How ORCA works under the hood: routing each request to the right agent, running generated code in an isolated sandbox, and turning repeat chat tasks into reusable workflows.

Agentic

The Studio

ORCA: one platform, many agents, working together

GovTech's MOM team built it to unify AI tools for analysing data, drafting slides and checking compliance

Agentic

The Lab

The Road Under the Harness

Previously I wrote about building a harness for yourself. This one is about the environment you're building in, and why at enterprise scale, if the platform underneath doesn't exist, individual wins have nowhere to accumulate.

AgenticInfrastructure