How We Turn Technical Demos into Must-Read Blog Posts with AI

As part of our How We Use Storytell to Build Storytell series, I’ll walk you through how I use Storytell to turn dense, technical engineering demos into blog posts for our website.

I'm Louise, Content Strategist at Storytell. Every week, I take our engineers' latest walkthroughs, often full of live builds, detailed updates, and internal shorthand, and turn them into something anyone can understand. Here's the system I’ve built to make that fast, accurate, and repeatable.

The challenge: translating "engineer-speak" into user-friendly stories

Our weekly engineering demos sessions are the lifeblood of our product development. But the language and framing are often geared toward an internal, more technical audience.

My job is to bridge the gap between that raw, unfiltered information and you, our users. I need to translate the demo’s rich context into something more structured, digestible, and aligned with our brand voice. On its own, that can be a painstaking process, but with the right setup in Storytell, it becomes something anyone can replicate. It’s not just about rephrasing; it’s about finding the story within the updates and guiding Storytell to help tell it clearly.

My workflow: from raw demo to polished draft in six steps

My process turns this complex translation task into a straightforward, repeatable workflow. It’s a system built on providing Storytell with high-quality context and guiding it with precise instructions. Here’s a detailed look at how it works.

Step 1: gathering the raw materials 

My work on a new demo blog begins before the demo even ends. During the live session, I take notes and timestamp important sections directly in our Notion agenda, marking key moments or features that I know will be important to highlight.

Once the demo is over, I collect all the source materials. This includes the full Zoom recording (which I grab from our Slack channel), the Otter transcript, and the Notion agenda from the meeting. Having these three distinct sources is critical, as each provides a different layer of context: the agenda gives structure, the transcript provides the raw text, and the video offers the visual and tonal nuance of the presentation.

Step 2: Creating a dedicated Collection 

Next, I create a new, dated sub-Collection within Storytell specifically for that week's demo. I then upload the transcript and the agenda document into this Collection. 

This is one of the most important steps in the entire process. By isolating the relevant documents in their own Collection and choosing that Collection as my preferred knowledge source, I'm creating a focused "brain" for Storytell. It won't get distracted by other information in our knowledge base; it will base its analysis exclusively on the specific materials I provided for this task, ensuring maximum relevance and accuracy.

Step 3: running the "engineering demo blog" prompt 

This is where the magic happens. I open SmartChat™ and use a detailed, multi-part prompt I’ve created and saved in our team’s Prompt Library called "Eng demo blog creation."

This is a comprehensive creative brief that guides Storytell with incredible specificity. The prompt instructs Storytell to:

• Act as an expert writer: It sets the persona and asks it to adopt Storytell's engaging, informative, and user-centric voice.
• Filter information carefully: This is a critical instruction for security and clarity. The prompt explicitly tells Storytell to only include updates that are marked as "Shareable to the public" in the agenda document. This automated filtering prevents any internal-only discussions or unreleased features from accidentally making their way into the public blog.
• Follow a specific structure: The prompt dictates the entire format: an introduction that enumerates the updates, a main body with clear, sentence-case headers for each feature, and a concluding summary.
• Adhere to stylistic rules: It specifies the word count and reminds Storytell to use our brand name "Storytell" correctly.

Here is the prompt I use:

You are tasked with creating an engineering demo blog post for Storytell based on the latest call with the engineering team. This blog will be published on the company website to inform users about new features and updates while subtly encouraging non-users to explore the product.

Instructions:
Review the following information:
- The file provided for the engineering call (@mention). This document includes key details on what the presentation is about, why it matters, additional context, and whether it is shareable publicly. Please do not include in the blog any details marked as “No” under the “Shareable to the public (Y/N)?” question. This also applies to similar sections from the transcript.
- The transcript of the call (@mention)

Write the blog post using the brand's voice (@mention), maintaining an engaging and informative tone.

Keep the content clear, descriptive, and focused on how users benefit from the updates.
Avoid assumptions about what users may feel—stick to concrete information.

Structure the blog post as follows:
Introduction: Set the context and highlight why these updates matter. Follow this format: 
“In this week’s engineering demo (DATE), our engineering team shared updates on…” (then enumerate the covered updates).
Main body: Use headers that clearly describe each update.
Conclusion: Summarize key takeaways.

For each feature or update discussed:
- Explain it in clear, informative language.
- Highlight the benefits users will get.
- Use paragraph formatting, with bullet points where helpful.
- Minimize sub-subheadings.
- When mentioning engineers who shared updates, include their job titles if provided in the Smart Brevity file.

Follow these writing and formatting guidelines:
- Use “Storytell” instead of “Storytell.ai.”
- Keep headers and subheaders in sentence case (only the first letter capitalized).
- Keep the language clear and concise.
- Create five alternative title options for the blog post. Ensure they align with previous engineering demo blog titles.

Word count: 800–1000 words.
  

Step 4: The human-in-the-loop review 

Within minutes, Storytell generates a complete first draft. My role now shifts from writer to editor and strategist. I create a Google Doc for the draft and begin the review process. While I read through the text, I play the demo recording on another screen, often at 1.75x or 2x speed, to ensure the draft accurately reflects the spirit and details of the presentation. This is where the human touch is invaluable. I check for flow, add nuance, and ensure the user benefits are crystal clear.

Step 5: enriching with visuals and links 

A blog post is more than just text. As I review the draft, I pause the video at key moments to take screenshots that visually explain the new features. I also add any necessary links to other articles or resources and trim the demo video itself—using tools like QuickTime—to create a concise cut for our YouTube channel.

Step 6: publishing and promotion 

Once the text is polished and the visuals are in place, I upload the final draft to Webflow, embed the newly trimmed YouTube video, and hit publish. But the work doesn't stop there. I then go right back into Storytell and ask it to generate social media posts based on the final blog content to share across LinkedIn, Facebook, and X. It’s a beautifully efficient cycle: Storytell helps create the core content, and then helps promote it.

Why this approach is a game-changer

This workflow has fundamentally changed how I approach content creation.

• It’s incredibly efficient: What used to consume half my day now takes about an hour and a half from start to finish. This frees me up to focus on higher-level strategy, planning, and other creative initiatives.
• It’s authentic and reliable: Because the draft is generated directly from the engineers' own words in the transcript and agenda, the content is grounded in truth. The AI is a tool for translation, not fabrication. The strict filtering instruction also provides peace of mind that we're sharing information responsibly.
• It’s scalable and consistent: This process isn’t a one-off trick; it’s a system. It works every single week, regardless of the complexity or number of updates. This ensures a consistent level of quality and a reliable stream of content for our users.

Ultimately, this workflow is a perfect example of human-AI collaboration. I don't need to be an engineer to communicate what our team is building. I just need to give Storytell the right context and the right instructions to bridge that knowledge gap for me.