3 Use Cases for AI in Sales Engineering

Can AI Help Sales Engineers Work More Effectively?


We asked Sales Engineering leaders from top SE teams at Snowflake, AT&T, and more about their uses for generative AI. This article condenses our findings to present three realistic use cases for AI that Sales Engineers are considering right now.


Sales Engineers (SEs) manage tons of information - ranging from the details of customer deployment to the technical inner workings of their products. They're an invaluable source of truth for other internal teams and their customers.

We spoke to 12 SEs from leading teams across the industry - they commonly described spending too much time on administrative tasks, leaving them too little time and energy to drive positive outcomes with customers. Thanks to Large Language Models (LLMs), SEs are rethinking the various types of "busy work" they spend time on. Can AI help?

#1. AI-Powered Knowledge Base


Organizing information into an easy-to-navigate knowledge base is essential to the smooth functioning of any Sales Engineering team. A knowledge base helps onboard teammates faster, allows you to deflect questions from account reps and support teams, and broadly ensures your team is on the same page.

Knowledge Bases Today

Knowledge bases for Sales Engineering teams commonly include onboarding and demo flows, technical documentation, system diagrams, sales collateral, product security information, and more.

As teams grow, these knowledge bases become outdated, difficult to navigate, and tedious to maintain. We frequently heard about the broken search experience in these tools. You need to know specific jargon and keywords to find what you need.

Popular Tools: Notion, Drive, Confluence, and Sharepoint.

AI can turn your Knowledge Base into an Answer Engine.

AI Search can turn your knowledge base into an interactive and intuitive search engine. You can not only find the information you need without knowing the exact keywords to search for - but you can also guide the formatting of the answer so that it's immediately useable.

Let's take the example of an SE named James, who wants to show a potential customer in the telecom industry examples of his success working with similar customers.

The Old Way
  1. James looks for case studies in his team's knowledge base by searching "case studies" and then "telecom," "telecommunications," "ROI," and other keywords in the search bar.
  2. He digs through several irrelevant documents until he finds ones that apply. He opens three relevant case studies in different tabs.
  3. From each page, he reads the document, copies useful sections, and pastes them into another document.
  4. He spends time synthesizing those sections into a format he can share via email.

The New Way
  1. James asks AI to "write three bullets about our successful outcomes working with other telecom corporations."
  2. AI provides an answer in seconds

LLMs with the proper retrieval infrastructure can allow you to quickly search through vast amounts of information and synthesize the findings into usable answers.

Must have features
  1. Accurate Search and Citations - The best AI search tools use proprietary models and advanced filtering and reranking techniques to ensure you get the correct answer. Look for accuracy benchmarks to ensure your solution doesn't hallucinate.
  2. Security - Connecting public LLM models directly to your internal data is a big no-no for data privacy and security reasons. See our post on Enterprise AI Adoption for more information on how you can protect your data from LLMs.
  3. Access Controls - You should ensure that permissions from source documents sync to your knowledge base. This ensures that answers your AI provides come from documents that the user asking the question has access to.
  4. Data Freshness - Data in your knowledge base must be verified consistently to ensure up-to-date and accurate answers.

#2. AI-Powered RFP and Security Questionnaire Responses


RFPs, RFIs, and security questionnaires can waste days of SE's time (and they're no fun to work on). While they're necessary for many sales processes - they're full of questions you've answered before and are usually a long, tiring game of search and copy-paste. Leveraging AI to automate this process can save Sales Engineers tons of time while removing a source of stress and boredom.

RFx Tools Today

Existing RFx tools rely on building an extensive Question-Answer Bank separate from your internal knowledge base. Your answer bank will frequently need to be updated. Any automation relies on exact matches between questions in the RFx and those in your bank, usually failing with even slight variations of the question. Finally, the answers lack personalization.

Popular Tools: Loopio, RFPIO, HyperComply

AI Can Respond to Hundreds of Questions in Minutes

LLMs are great at reading large documents, sifting through tons of data, and finding the answer regardless of the question's phrasing. They can also provide deep personalization of the response to tailor content to each opportunity.

Let's see our hypothetical SE James respond to a security questionnaire.

The Old Way
  1. James reads through a questionnaire and kicks off the response process with stakeholders in sales, security, and product. He spends time divvying up questions amongst the team.
  2. He writes an answer by hand in an excel spreadsheet for questions he's very familiar with.
  3. For others, he digs through his knowledge base and previous responses the team has filled out. He manually customizes the response to fit the context of the particular question and the RFI he's answering.
  4. It takes James and the team 13 hours of work across three days to complete the RFI from end to end.

The New Way
  1. James uploads the RFI and receives a first draft response from the AI within 5 minutes. The AI finds answers from old RFIs and James' knowledge base.
  2. James reviews the answers over an hour, messaging teammates for input only when he needs clarification.
  3. It takes James 2 hours of work and one day to complete the RFI from end to end.

Must-Have Features
  1. Flexible Document Ingest - RFPs and Security Questionnaires come in various formats based on the buyer (spreadsheet, document, web portal, etc.). Your solution should allow you to respond in any format quickly.
  2. Answer Personalization - Each opportunity and prospective buyer is different. You'll often need to personalize answers to match the buyer's needs, budget, and team. Your automation tool should refrain from providing boilerplate answers that will lose you opportunities.
  3. Human Review Workflows - With any AI-generated content, you want your software to introduce intentional friction so that a human has to review and approve the content. Your tools should also allow SEs to notify SMEs when they need assistance and track due dates.
  4. Citations - Each answer should be reproducible and traceable via citations and links to source materials. This functionality makes it easy to verify answers and finalize your response submissions.

#3 Conversational Intelligence Workflows


Conversational Intelligence tools help SEs understand their team's discussions with specific customers, enabling them to propose customized solutions. This data is often used to write proposals, prepare for meetings, and customize deal room documents. LLMs are a powerful method for integrating the knowledge gained from meetings into sales workflows and automating these workflows.

Workflows Today

The workflows powered by conversational tools like Gong are built for sales reps - they can help write a better email or perform better in their next discovery call. Sales Engineers tend to use the data in these tools as guidance while manually completing the workflows mentioned above.

Popular Tools: Gong, Chorus, Firefiles.ai, Otter.ai, and Fathom

Powering Sales Engineering Workflows with AI

AI can use data in these platforms to inform the customization of proposals, making them relevant and compelling with little human intervention. AI can also emphasize specific phrases or value propositions that resonate with the customer, which can be integrated into the proposal to enhance effectiveness.

AI can also help in preparing for follow-up meetings or presentations. It can provide a comprehensive analysis of previous interactions with the customer, highlighting the main topics of interest or concerns. This preparedness can make the proposal defense more successful and further improve the chances of closing the deal. AI's ability to automate these tasks lets sales engineers focus more on building stronger customer relationships and perfecting their sales strategy.

Must-Have Features
  1. Security - Allowing AI interactions with sensitive company systems can be risky. With proper guardrails (for example, only "Read" operations), teams can limit risks.
  2. Traceability - You'll need to know which data was used to inform each section of an AI-generated document. This helps you validate that the content is correct and reliable.


While AI has the potential to revolutionize Sales Engineering workflows, it's essential to understand its limitations and implement realistic solutions that speed up your processes instead of slowing you down. With effective use of large language models, Sales Engineers can automate time-consuming tasks, provide personalized content, and increase the utility of knowledge bases. By doing so, they can work more efficiently and effectively, bringing in more revenue for their organizations.

If you're interested in learning more, check out www.dealpage.ai. Our mission is to make Sales Engineering easier through the power of AI and intelligent automation.