From Development to Production: Ask ServiceNow's AI Experts Anything

[ryanyong]

Hey everyone, I'm Ryan, an engineer on ServiceNow's Forward Deployed Engineering team. We're a team of experts and builders who've gained deep AI deployment expertise at tier 1 technology companies and research labs. We're certified experts on the ServiceNow AI Platform, with direct access to Product and the ServiceNow C-suite.

What we do
We partner directly with customers facing the most complex challenges to co-innovate agentic AI use cases and ship them to production on the ServiceNow AI Platform.

Why this thread
If you're a ServiceNow customer sitting on AI entitlement you haven't fully put to work, this is your chance to ask the Forward Deployed Engineers building these solutions in the field. Bring your hardest questions about moving from unused credits to AI running in production, and getting the most value out of your existing ServiceNow investment.

Ask us anything. We are the subject matter experts, and we will get you an answer.

We'll be answering live from August 3rd to 7th. Post your question anytime between now and 8/7, upvote the ones you want answered first.

[Dustin Gannon]

I would guess you want to dig into your ai search profiles and make sure they are correctly setup and have the filtering criteria you want.  This area of the system has a decent amount of complexity and it can be tricky to get all the records correct.  The new AI Search Preview helps show this information too if you are using the older view.  Make sure you have that installed.

[Christine Kafka]

What are the best practices for training/testing AI Agents and how do you migrate that training to other environments, including production.

[NicholasG799675]

The best practice is to treat AI agents like enterprise software instead of one off experiments. Define success criteria before development, build a robust evaluation framework using real-world scenarios, separate deterministic business logic from AI reasoning, and validate data quality before validating the agent.

For migration, use the same disciplined promotion process you would use for any business-critical application: Development → Test/UAT → Production, with regression testing, version control, governance reviews, and measurable acceptance criteria at every stage.

Most successful deployments don't improve because the model gets smarter but instead because the data, workflows, development processes mature over time.

[IntoTheBreach]

Why does the Now Assist Virtual Agent capability supersede established Virtual Agent (VA) capabilities? It appears we have to either custom code our way around the Now Assist so our preprogrammed VA processes go first or we have to consume Assists to leverage work we have already done.

[ryanyong]

VA relies on NLU (Natural Language Understanding), whereas NAVA relies on LLMs. Given the additional flexibility and stronger OOTB performance, NAVA supersedes VA. To go from VA to NAVA, ServiceNow has a topic migration tool in Virtual Agent Designer to convert the legacy topics into NAVA-friendly topics, which should prevent needing to consume Assists to do previously completed work. Docs here.