Organizations across every industry are exploring ways to redesign work, improve efficiency, and responsibly adopt AI. One of the fastest-growing areas of transformation is the rise of AI agents — or as we like to call them, AI teammates.
To help people understand what’s possible and how to get started, we hosted a live workshop focused on building and deploying AI agents using no-code tools like Microsoft Copilot Studio. Many participants were new to agent building, so the session walked step-by-step through the foundations, the technology, and the real-world impact of introducing AI automation into HR and operational workflows.
Below is a comprehensive recap with the key insights, examples, and best practices we covered.
One of the biggest misconceptions about AI is that it replaces entire roles. In reality, AI agents automate tasks, not skills. They act as digital teammates who can carry out repetitive, rule-based, time-consuming work — freeing humans to focus on strategic, creative, and relational tasks.
During the session, Siobhan Savage, CEO & Co-Founder of Reejig, emphasized that AI agents are now a critical capability for businesses preparing for the future of work. She highlighted increasing customer demand for use cases like a “Bonus Buddy” AI agent that handles bonus-related HR questions — an example with clear ROI and immediate practicality.
She also noted an important industry trend:
Microsoft is quickly becoming the dominant front-end for enterprise AI, while other vendors will increasingly serve as core data sources rather than delivery layers. This clarity matters for leaders mapping their AI strategy.
The session included a live demonstration showing how easy it is to create an AI agent without writing a single line of code. Using Microsoft Copilot Studio, Mike walked through the entire lifecycle of building an AI teammate called “Bonus Buddy.”
1. Creating the foundation in Microsoft Teams
> Opening Microsoft Copilot Studio directly inside Teams
> Setting the agent’s purpose: answer bonus policy questions
> Applying restrictions to prevent access to PII
> Ensuring the agent only uses approved data sources
2. Adding HR policy knowledge using SharePoint
A key best practice for AI agents is to avoid static uploaded files. Instead, SharePoint links allow real-time policy updates, meaning the agent always references the most current version of HR and bonus documentation.
The agent’s knowledge base included:
> Bonus policy overviews
> FAQs
> Eligibility rules
> Internal HR support guidelines
> Additional documentation needed to answer edge cases
3. Training the AI agent
We explored what makes prompt engineering effective today — and why the skill will become less critical as agents evolve.
To improve accuracy, we demonstrated how to:
> Write clear, specific prompts
> Use the tool itself to generate prompt drafts
> Set strict context boundaries to prevent hallucinations
> Ensure the agent follows policy documentation word-for-word
4. Deploying the AI agent into real workflows
Deployment is where the agent becomes a true AI teammate.
Mike demonstrated how to:
> Publish the agent into a Microsoft Teams channel
> Integrate with Slack for cross-platform communication
> Track trends and insights from employee questions
This showcased how no-code AI tools can support both employees and leaders by surfacing what people repeatedly ask about — creating a feedback loop for continuous improvement.
Throughout the session, Siobhan emphasized that responsible AI adoption requires intentional design, not just good technology.
Key guidelines include:
✔ Start with a valuable, real-world use case: Choose a problem that employees experience regularly, such as navigating HR policies.
✔ Use dynamic, live documents for knowledge: SharePoint links ensure your AI agent updates automatically as policies change.
✔ Test edge cases extensively: Your agent should handle the difficult, ambiguous questions — not just the easy ones.
✔ Deploy inside platforms employees already use: Microsoft Teams and Slack drive faster adoption and higher trust.
✔ Avoid PII and follow enterprise data governance: Safe AI requires clarity on what the agent can and cannot access.
An important question about governance models was raised: How do companies balance centralized control with the flexibility needed for innovation?
Siobhan emphasized that AI agents are not just a technology implementation — they require:
> Workflow redesign
> Clear communication
> Training and expectation setting
> Guardrails that support safe experimentation
The conversation also highlighted personal applications of AI agents, encouraging people to use AI to streamline daily tasks outside of work so they can build confidence and capability.
AI agents are quickly becoming a foundational component of modern workforce strategy. The session reinforced that with the right approach, anyone can build an AI teammate that delivers real operational value, improves employee experience, and reduces administrative burden.
We will continue developing resources such as:
> A user flow playbook for building AI agents
> Future learning sessions to deepen technical and strategic skills
> Templates for common HR and business agent use cases
If your team needs help spotting the right tasks ready for automation or augmentation, get in touch with your team.