Have you integrated Bob with any AI-friendly tools?

Wondering if you’ve integrated with any AI tools yet?

 ​@Megan Baldock, ​@Sophie Johnson, ​@Catherine Hitchins, ​@Taylor Lowe, ​@Augustas Baltrenas, ​@Denise Vuilleumier, ​@Carla Romero, ​@Dina Arni 

We have not yet, but are starting to talk about options.  I’ll be curious to see if others have.

We haven’t integrated specific LLMs, but we’ll be looking to building AI agent and connecting HiBobs MCP to it. However, we currently are building some integrations with n8n which has built-in AI agent nodes and LLMs.

We are working on an LLM that can score resumes for us. The AI resume tool is great, but when you have thousands of resumes to go through, it’s still time consuming to toggle between the resume and the AI summary. Our data team built us an external tool that will score the candidates on a scale of 1 to 10

We have not integrated yet. I’m excited for the new Bob chat tool though. I’ve also used the AI tools built directly into Bob for Analytics, Talent, and Learning quite a bit though!

​@Megan Baldock, I’m curious about this-and if you have any ideas on how to approach compliance like the one below for Illinois or Colorado. We are big on AI but haven’t maximized our use of Bob yet which we’re hoping to work on soon.

My coworker shared this: 

AI Compliance: What's Changed in 2025?

The past year brought an unprecedented wave of AI-related policy developments:

• EU AI Act: Went into force in August 2024, with enforcement beginning in August 2026. It introduces a tiered risk model, where systems categorized as high-risk must adhere to strict requirements around documentation, transparency, and human oversight. This applies to everything from biometric identification systems to credit scoring algorithms.
• Colorado AI Law: This law takes effect in February 2026 and mandates impact assessments for AI systems that significantly affect consumers. It explicitly prohibits algorithmic discrimination, making fairness testing and documentation non-negotiable for companies operating in or serving users in Colorado.
• Illinois HB 3773: Coming into force in August 2025, this regulation targets employer use of AI in hiring and workforce management. It imposes requirements for fairness audits and explanations for AI-driven employment decisions.
• U.S. Executive Order on AI (Oct 2023): This executive action kickstarted regulatory activity across multiple federal agencies. The National Institute of Standards and Technology (NIST), the Federal Trade Commission (FTC), and the Department of Homeland Security (DHS) have since released frameworks and guidance for risk management, transparency, and accountability.

Taken together, these laws and policies now form the backbone of global AI compliance expectations. They are reshaping how AI is deployed, governed, and audited. 

EU Artificial Intelligence Act | Up-to-date developments and analyses of the EU AI Act

A post on EU Artificial Intelligence Act provided by: https://artificialintelligenceact.eu

Hi ​@HollieLouiseMann, ​@Lucy Taylor, ​@Johnny Barnett, ​@Marina Petrovska - Wondering if any of you have explored any AI integrations yet?

Are there any product updates in scope to enhance the AI chatbot feature on Bob’s homepage to use in the same way as the MCP tool?

Some other systems have this feature in built where employees and managers can ask questions about data such as ‘Who is XX’s manager’, etc.

It would be great if this was part of Bob without the need to connect to other AI tools.

​@Karen Stewart Longest Right now, we are only hiring in NYC. But this is definitely something that is in our minds as we move forward as I’m sure we will continue to see more states adopt similar rules. 

​@Megan Baldock, that’s what we think too. I’m actually surprised NYC doesn’t have one already but I know that NY has one in the state senate. I’m also curious how to prevent any issues like discrimination using AI.  

On the compliance topic I think it’s beneficial that HiBob offers decent APIs and is experimenting with an official MCP server, while giving customers the freedom to connect any AI/LLM with HiBob.

This gives us the option to choose a model that is covered by our existing contracts and legal frameworks. As a real-world example: Most of our cloud infrastructure already runs on Microsoft Azure, we use what Microsoft calls “Azure AI Foundry Models”. These are basically OpenAI models that do not run on standard OpenAI servers, but rather in an Azure data centre that has been approved for processing PII.

Other data centre operators offer similar services for other major AI providers (for example, you could run Claude through Amazon’s AWS). For highly sensitive use cases, it is even possible to use a local model running in your office or private data centre.

While this approach requires slightly more effort, the privacy benefits and flexibility it offers far outweigh the drawbacks.

​@tom, I’m really appreciative of this post- love the emphasis on flexibility and independence in your approach. We run a lot on Azure, so Microsoft tools are a natural fit for us. Any tips on where to start if we want to explore connecting our own models?

Thank you for sharing ​@tom! This is awesome-already shared with a few friends and will share with a few more when they’re back from vacay. 

Hello Augustas, that's an interesting question!

1. Was the AI agent created in ChatGPT? How did you connect it to HiBob?

2. Are you connecting to N8N through API keys?

If you could help me with that, I'd really appreciate it.

Will you add n8n for AI automation workflows with HiBob?

Hi ​@Augustas Baltrenas Tagging you in, so you don’t miss out on this! :) 

Hi,
We haven’t yet explored separate AI agent creation and connection with HiBob through MCP. 
However, yes, we are connecting HiBob with n8n via HiBob’s Public API and n8n’s HTTP request node since HiBob doesn’t have built-in integration with n8n. In a way you can create autonomous AI agent that performs tasks this way it’s just limited with what endpoints HiBob’s API offers, but mostly useful for creating different automations without the help of IT department and without coding in general.

Super interested in hearing from and learning from people who have connected Bob with AI tools (we use CoPilot) via MCP. 

One of the use cases I would want to use it for (if possible) is to set up self-serviced reporting for Finance teams etc., directly via CoPilot and not through building dashboards in Bob. 

I would also love to use it to scan Bob for not-up-to-date-people-data from employee profiles and send teams messages directly to employees and/or managers to get them to update it, to ensure that we capture 100 % of the employee data we need.