Turn Your Project Files Into an AI Knowledge Base

🧭 THIS WEEK AT AI SECOND ACT

Howdy, this week it’s a quick one! Building upon last week where we discussed projects vs custom GPTs, we’ll add knowledge and context through files. Just some of the key advantages:

• Instant summaries of large documents

• Fast and specific information retrieval across multiple documents

• Context-specific problem solving & analysis

If you’re enjoying this newsletter, please share it! Forward this to a friend/colleague and have them subscribe at aisecondact.com/subscribe. Thanks!

👉 You can also just hit "Reply" and let me know what you want more (or less) of or use the poll to give feedback.

My goal is to make this as valuable and practical as possible as we navigate the new AI era. 🚀

🧰 AI NEWS + LEARNING

Here’s a few things I found recently:

• How AI Is Helping Students Find the Right College

• Microsoft Copilot Prompt Gallery

• AI Fluency: Frameworks and Foundations by Anthropic [maker of Claude] —> I plan to do this myself

• aiEDU —> The AI Education Project —> preparing kids & educators for AI - very cool and possibly a focus of AI Second Act in the future - is this interesting to you? Hit Reply!

• Generative AI for Beginners - Free Microsoft 18 lesson comprehensive course!

🗺️ FEATURED INSIGHT

🧰 THE FILE UPLOAD GAME-CHANGER

Ever wish your AI assistant just knew your project without 50 prompts or re-entering information/knowledge? Now it can.

What changed: Every major AI tool now supports multiple file types, PDFs, spreadsheets, images, code files, presentations, and more.

Why it matters: Instead of re-explaining your context every time, you upload your files once and AI "remembers" everything within the same project thread or a custom GPT you create.

The unlock: Your messy project folder becomes a searchable, analyzable knowledge base. Like having a research assistant who's already read every doc in your cabinet.

📊 What You Can Upload (By Tool)

🗺️ 5 MINI KNOWLEDGE BASES THAT WORK

1. Project Context Folder
Upload: Project charter, meeting notes, status reports, budget spreadsheets
Ask: "What are the biggest risks in this project and what haven't we considered?"
Why it works: AI connects dots across status, risks, budgets, and decisions—fast.

2. Competitive Intelligence Pack
Upload: Competitor websites (as PDFs), industry reports, pricing sheets
Ask: "Where are we vulnerable and what opportunities are we missing?"

3. Performance Review Prep
Upload: Last 6 months of emails, accomplishments doc, goal tracking sheet
Ask: "Draft my self-review highlighting data-driven wins and strategic impact"

4. New Hire Onboarding
Upload: Team org chart, process docs, recent project summaries
Ask: "Create a 30-60-90 day plan for our new team member"

5. Budget Planning Base
Upload: Last 3 years of budgets, current spend reports, upcoming project requests
Ask: "What budget patterns should worry us and where can we reallocate?"

In summary, uploading files gives your chat a context and knowledge base. This gives you the ability to:

• Summarize documents or pull specific info instantly

• Analyze, research, retrieve information, and problem-solve using your files on top of the model’s existing general knowledge

⚡ PRO TIPS THAT SAVE HOURS

✅ Upload screenshots of dashboards (metrics, analytics, project boards)
✅ Include old email threads as text files—gives AI full context of decisions made
✅ Use descriptive filenames (e.g., "Q2_Sales_Analysis_2025.pdf" not "Document1.pdf")
✅ Upload similar documents together within a project or custom GPT—for stronger pattern recognition
✅ Use follow-up prompts like "Summarize this like an executive brief" or "Give me 3 key insights in bullet form"
❌ Don't upload sensitive data (passwords, SSNs, confidential client info)—stick to work you'd share with a colleague

🚀 ACTION CHALLENGE: Build Your First AI Knowledge Base

This Week's Challenge:

1. Pick one ongoing project or knowledge area you're working on

2. Gather 3–5 relevant files (PDFs, spreadsheets, presentations)

3. Upload them to ChatGPT or Claude into a project (or custom GPT if using ChatGPT)

4. Ask what you need:

   • Summarize large document XYZ

   • What is duplicate information between these files / how could we improve them or create a single file that captures all information better?

   • What are the basic features of product ABC?

Time investment: 10 minutes setup
Payoff: Hours of analysis you'd never have time to do manually

Real Example for a Product Manager:

• Upload press releases from competitors to summarize and store as a knowledge base

• Upload supplier technical documents for quick summaries and information retrieval

• Upload past roadmaps to brainstorm future direction or identify feature trends

Hope you understand the value of using files with the chatbots. Let me know if questions!

Future-proof your career with AI
— Brett

👉 Hit “Reply” and share your experience, let me know what you want — I read every one!

_{Picture by Fabian Barral on Unsplash.}