Salesforce is one of the fastest innovators in the SaaS AI space. They started with a prompt builder early in 2024. During Dreamforce they launched Agentforce to build your own automated AI Agents. In February of 2025 major expansions are planned, adding voice capabilities to agents, coaching your sales team with an agent and the expansion to many more countries. How can Salesforce move so fast and make sure its AI doesn’t hallucinate?
Subscribe to Techzine Talks on Tour and listen to our other episodes via Spotify, Apple , YouTube or another service of your choice.
For this episode of Techzine Talks on Tour, we talk to Adam Evans, SVP of Product for the Salesforce AI Platform. Salesforce combines many data sources with the Salesforce Data Cloud. He explains how the data is being used and how the reasoning engine loops through the data and AI queries to ensure the best consistent outcome is served to the end-user.
Salesforce has built a reasoning engine that handles data and AI queries. When a query is sent to an AI modal, it doesn’t always give back the right answer. Sometimes, it needs more information. The reasoning engine acts like a human to catch that. It will then search for more relevant information and feed it to the AI model. When you ask a question to a Salesforce AI Agent it may go back and forth several times to an AI model before you get an answer. Just to make sure you get the best possible answer.
Salesforce creates prompts on the fly and uses RAG to feed the most relevant information to the AI models. The reasoning engine works together with the AI Trust Layer to stay compliant and mask customer data. The reasoning engine keeps the AI on a leash.
The reasoning engine and the AI Trust Layer are all agnostic. With new AI models being released daily and the LLMs making giant steps forward every few months, it’s hard to tell which AI model will dominate the industry in 12 or 24 months. Salesforce is taking that into account with how it has built its AI framework; they can switch to different LLMs very easily.
Evans also addresses critical issues in AI data privacy, emphasizing the importance of data residency and localized computing to comply with international regulations. This episode is a must-listen for anyone looking to understand the future of AI technology in SaaS and how it can improve your business.
Where to find Techzine Talks on Tour?
Techzine Talks on Tour is available on all the well-known platforms. So you can also find it on Spotify, Apple Podcasts and YouTube for example. Just search for Techzine Talks on Tour in your favorite podcast app.
Previous episodes of Techzine Talks on Tour:
- NetApp solves challenges to data infrastructure at the platform level
- Unlocking AI in SaaS: Brian Chess on the transformation of NetSuite
- The SOC of the future: what is it and is it for everyone?
- Threat hunting is very important, but also very frustrating: how can AI help?
- How do you innovate for a future you can’t entirely predict?
Get in touch
We hope you like this new podcast series. If so, please let us know. If you have suggestions on how we can improve, we would like to hear those too. We’re also open to suggestions around specific topics, or specific people that want to appear in an episode of Techzine Talks on Tour. You can find both Coen and Sander on LinkedIn, or you can send an email to info@techzine.eu.
Also read: Salesforce raises the AI bar with Agentforce; are AI agents replacing employees?