AI-Powered Productivity Stack
Deep dives into Claude Projects, ChatGPT, Notion AI, Perplexity, Gamma, and more.
Building Your AI-Powered Toolkit
The AI landscape has matured significantly. Rather than relying on a single all-purpose tool, the most productive professionals use a curated stack of specialized AI tools — each chosen for what it does best. This module walks you through the leading AI productivity tools, explains their strengths, and helps you build a workflow stack tailored to your needs.
Claude Projects
Claude, made by Anthropic, offers a feature called Projects that lets you organize your AI usage into persistent workspaces. Each project maintains its own context, custom instructions, and uploaded knowledge base, making Claude significantly more useful for ongoing work.
Key Features
- Custom instructions: Set project-specific system prompts that shape how Claude responds within that project. For example, a "Marketing Copy" project could instruct Claude to always write in your brand voice and follow your style guide.
- Knowledge base uploads: Upload documents (PDFs, text files, code, spreadsheets) that Claude can reference across all conversations in the project. This is Claude's way of giving the model persistent context about your work.
- Conversation organization: Keep related conversations grouped together. Your research project conversations stay separate from your email drafting conversations.
- Shared projects: Share projects with team members so everyone benefits from the same custom instructions and knowledge base.
How to Set Up a Claude Project
- Open Claude at claude.ai and click on "Projects" in the sidebar
- Create a new project and give it a descriptive name
- Add custom instructions that define how Claude should behave within this project (tone, format, constraints, domain knowledge)
- Upload relevant documents to the knowledge base — style guides, product docs, research papers, or reference material
- Start a conversation within the project; Claude will automatically use your instructions and reference your uploaded documents
Claude by Anthropic
Anthropic
Claude's web interface with Projects, knowledge base uploads, and custom instructions for organized, context-rich AI conversations.
ChatGPT Custom GPTs
OpenAI's ChatGPT offers Custom GPTs — personalized versions of ChatGPT that anyone can create and share. A Custom GPT combines a system prompt, uploaded knowledge files, and optional tool integrations (web browsing, image generation, code execution) into a single shareable package.
Key Features
- Custom instructions and persona: Define exactly how the GPT should behave, what it should and shouldn't do, and what persona it adopts
- Knowledge files: Upload documents the GPT can reference, similar to Claude's knowledge base
- Built-in tools: Enable web browsing, DALL-E image generation, and Code Interpreter (for running Python code and analyzing data)
- GPT Store: Browse and use Custom GPTs created by others, or publish your own to the store
- Sharing: Share Custom GPTs via link with your team or make them public
Custom GPT vs Claude Project: When to Use Which
| Feature | Claude Projects | ChatGPT Custom GPTs |
|---|---|---|
| Best for | Ongoing work with evolving context | Shareable, reusable AI configurations |
| Knowledge base | Upload documents per project | Upload files per Custom GPT |
| Tool integrations | Web search, analysis tool, artifact creation | Web browsing, DALL-E, Code Interpreter |
| Sharing | Team sharing within projects | Link sharing or public GPT Store |
| Long-form writing | Stronger for long, nuanced documents | Good for shorter, structured tasks |
ChatGPT by OpenAI
OpenAI
ChatGPT with Custom GPTs, GPT Store, image generation, code execution, and web browsing capabilities.
Notion AI
Notion AI integrates AI capabilities directly into the Notion workspace — the popular tool for documents, wikis, databases, and project management. Rather than switching to a separate AI tool, you can invoke AI assistance right where you're already working.
Key Capabilities
- Writing assistance: Draft, edit, expand, shorten, or change the tone of content directly within Notion pages. Highlight text and ask the AI to improve it.
- Summarization: Summarize long meeting notes, documents, or database entries with a single click
- Brainstorming: Generate ideas, outlines, and creative content within your existing documents
- Q&A across your workspace: Ask questions about information stored anywhere in your Notion workspace, and the AI searches across your pages, databases, and documents to find answers
- Autofill database properties: Automatically fill in database fields based on page content — for example, extracting action items or categories from meeting notes
- Translation: Translate content into other languages directly in place
Notion AI
Notion
AI-powered writing, summarization, Q&A, and database automation built directly into the Notion workspace.
Perplexity
Perplexity is an AI-powered research tool that searches the web in real time and provides answers with inline citations. Unlike traditional AI chatbots, Perplexity is designed specifically for research and information gathering — every claim comes with a source you can verify.
Key Features
- Real-time web search: Perplexity searches the internet live to answer your questions with up-to-date information, not just training data
- Inline citations: Every factual claim includes a numbered source link so you can verify the information
- Follow-up questions: Perplexity suggests related questions to help you dive deeper into a topic
- Focus modes: Target your search to specific sources — academic papers, Reddit discussions, YouTube videos, or general web
- Collections: Organize your research into topic-based collections that persist across sessions
- File upload and analysis: Upload documents and ask Perplexity to analyze, summarize, or answer questions about them
When to Use Perplexity vs a General AI Chatbot
Use Perplexity when you need:
- Current information that may have changed after AI training data cutoffs
- Verifiable facts with citations you can check
- Research across multiple sources on a topic
- Academic or scientific literature review
- Competitive intelligence or market research
Use Claude or ChatGPT when you need:
- Creative writing, brainstorming, or content generation
- Long-form document creation or editing
- Complex analysis of your own data or documents
- Code generation or technical problem-solving
Perplexity
Perplexity AI
AI-powered research engine that searches the web in real time and provides answers with inline citations from verified sources.
Gamma
Gamma is an AI-powered tool for creating presentations, documents, and web pages. Instead of wrestling with slide layouts and design choices, you describe what you want and Gamma generates polished, visually appealing content.
Key Features
- AI-generated presentations: Describe your topic and Gamma produces a complete slide deck with content, layout, and visuals
- Multiple formats: Create presentations, documents, or web pages from the same content
- Smart design: Automatically handles layout, spacing, typography, and image selection so you can focus on content
- Edit and refine: Adjust individual slides, swap themes, change images, or ask the AI to rewrite sections
- Import content: Paste in notes, documents, or outlines and have Gamma transform them into a presentation
- Sharing and analytics: Share via link and track views and engagement
Gamma
Gamma
AI-powered tool for creating presentations, documents, and web pages with automatic design and layout.
Otter.ai
Otter.ai is an AI-powered meeting assistant that joins your calls, transcribes conversations in real time, and generates summaries and action items. It integrates with Zoom, Google Meet, and Microsoft Teams.
Key Features
- Live transcription: Real-time transcription during meetings with speaker identification
- Automated summaries: AI-generated meeting summaries highlighting key decisions, action items, and discussion topics
- Action item extraction: Automatically identifies and lists action items from the conversation
- Chat with your meetings: Ask questions about past meetings and Otter will search through transcripts to find answers
- Calendar integration: Automatically joins scheduled meetings when connected to your calendar
- Shareable notes: Share meeting notes with attendees (and non-attendees) via link
Getting the Most from Otter
- Connect Otter to your calendar so it automatically joins and records meetings
- After each meeting, review the AI summary and action items — edit anything the AI missed or got wrong
- Use the "chat with your meetings" feature to recall decisions from past meetings without re-reading full transcripts
- Share summary notes with stakeholders who couldn't attend
- Use transcripts as input for Claude or ChatGPT to generate follow-up emails, project updates, or documentation
Otter.ai
Otter.ai
AI meeting assistant with live transcription, automated summaries, action item extraction, and meeting Q&A across Zoom, Meet, and Teams.
NotebookLM by Google
NotebookLM is Google's AI research tool that lets you upload sources — documents, web pages, YouTube videos, and audio files — and then chat with an AI that is grounded exclusively in those sources. Unlike general AI chatbots, NotebookLM only answers based on the materials you've provided, which makes it excellent for research where accuracy and source attribution matter.
Key Features
- Source-grounded AI: Responses are based solely on your uploaded sources, reducing hallucination. Every claim can be traced to a specific source.
- Multiple source types: Upload PDFs, Google Docs, Google Slides, web URLs, YouTube videos, and audio files as sources
- Audio overviews: NotebookLM can generate podcast-style audio summaries of your sources — a feature that gained significant popularity when it launched
- Notebook guides: Auto-generated study guides, FAQs, timelines, and briefing documents based on your sources
- Inline citations: Click on any claim to see which source it came from, with direct quotes highlighted
NotebookLM
Google's AI research tool that lets you upload sources and chat with an AI grounded exclusively in your materials, with inline citations and audio overviews.
Building Your Personal AI Workflow Stack
The goal isn't to use every tool — it's to build a minimal, effective stack that covers your core needs. Here's a framework for choosing:
The Core Stack (Start Here)
Primary AI Assistant
Choose one as your main tool for writing, analysis, and general tasks: Claude or ChatGPT. Use Projects or Custom GPTs to customize it for your recurring workflows.
Research Tool
Perplexity for real-time web research with citations. Use NotebookLM when you need to analyze a specific set of documents in depth.
Meeting Assistant
Otter.ai for transcription, summaries, and action items. Set it up once and it works automatically for every meeting.
The Extended Stack (Add as Needed)
Presentations
Add Gamma if you create presentations regularly. Draft content in your primary AI assistant, then use Gamma for design and layout.
Workspace AI
If your team uses Notion, enable Notion AI to get writing assistance and Q&A directly in your existing workspace without switching tools.
Workflow Templates
Here are ready-to-use workflows that combine multiple tools for common tasks.
Workflow 1: Research Report
Step 1: Discovery (Perplexity)
Search for the topic on Perplexity. Collect key sources, statistics, and viewpoints. Save to a collection.
Step 2: Deep Analysis (NotebookLM)
Upload the most important sources to NotebookLM. Use it to cross-reference, find connections, and generate a source-grounded summary.
Step 3: Draft (Claude)
Use Claude with a project set up for your report's domain. Provide the research from steps 1 and 2 as context. Draft each section using prompt chaining.
Step 4: Presentation (Gamma)
Take the executive summary and key findings into Gamma to create a presentation for stakeholders.
Workflow 2: Meeting Follow-Up
Step 1: Record (Otter.ai)
Let Otter join the meeting automatically. Review the AI-generated summary and action items after the call.
Step 2: Process (Claude)
Copy the Otter transcript into Claude. Ask it to: draft a follow-up email to attendees, create a project update for stakeholders, and generate a task list with owners and deadlines.
Step 3: Document (Notion)
Save the meeting notes and action items to Notion. Use Notion AI to update any related project documentation or databases.
Workflow 3: Content Creation Pipeline
Step 1: Research (Perplexity)
Research the topic for your content. Gather statistics, quotes, and existing perspectives from credible sources.
Step 2: Outline and Draft (Claude)
Create a Claude Project for your content brand. Upload your style guide and past content as knowledge base. Use prompt chaining: first generate an outline, then draft each section, then review and refine.
Step 3: Repurpose (ChatGPT + Gamma)
Use ChatGPT to repurpose the article into social media posts, email newsletter copy, and talking points. Use Gamma to create a visual slide version for LinkedIn or presentations.
Workflow 4: Weekly Planning and Preparation
Step 1: Review (Otter.ai + Notion)
Review Otter transcripts from the past week's meetings to identify open action items and commitments. Check Notion for outstanding tasks.
Step 2: Plan (Claude)
Use Claude to prioritize your tasks for the week. Provide your goals, deadlines, and constraints. Ask for a suggested daily schedule with time blocks.
Step 3: Prep (Perplexity + Claude)
For each important meeting, use Perplexity to research attendees or topics, then use Claude to prepare talking points, questions, and agendas.
Quick Reference: Tool Comparison
| Tool | Best For | Pricing | Key Differentiator |
|---|---|---|---|
| Claude | Long-form writing, analysis, coding, structured work | Free tier + Pro plan | Projects with persistent context, large context window |
| ChatGPT | General tasks, image generation, code execution | Free tier + Plus plan | Custom GPTs, DALL-E, GPT Store ecosystem |
| Notion AI | In-context writing and workspace Q&A | Included in Business and Enterprise plans | Embedded in your existing workspace |
| Perplexity | Research with real-time web search and citations | Free tier + Pro plan | Real-time search with inline source citations |
| Gamma | Presentations and visual documents | Free tier + paid plans | AI-generated visual design and layouts |
| Otter.ai | Meeting transcription and summaries | Free tier + paid plans | Automated meeting join, transcription, and action items |
| NotebookLM | Source-grounded research and analysis | Free | Answers grounded exclusively in your uploaded sources |
Resources
Claude
Anthropic
Anthropic's AI assistant with Projects for persistent context, knowledge base uploads, and long-form document capabilities.
ChatGPT
OpenAI
OpenAI's AI assistant with Custom GPTs, image generation, code execution, and a marketplace of community-built GPTs.
Perplexity
Perplexity AI
AI-powered research engine with real-time web search, inline citations, and focus modes for targeted research.
NotebookLM
Google's source-grounded AI research tool with document upload, inline citations, and audio overview generation.
Key Takeaways
- 1Build a minimal AI stack: one primary assistant (Claude or ChatGPT), one research tool (Perplexity), and one meeting tool (Otter.ai). Add more only as needed.
- 2Claude Projects and ChatGPT Custom GPTs both let you create persistent, customized AI configurations — choose based on your primary use case and preferences.
- 3Perplexity is purpose-built for research with citations; NotebookLM excels at analyzing specific documents you upload. They complement each other.
- 4Notion AI is most valuable when your team already uses Notion — it brings AI assistance directly into your existing workflow without tool-switching.
- 5Gamma handles the visual design of presentations so you can focus on content. Draft in your AI assistant first, then design in Gamma.
- 6The most productive AI workflows chain multiple tools together: research in Perplexity, analyze in Claude, present in Gamma, document in Notion.
- 7Start with 2–3 tools, master them, then expand. Tool overload reduces productivity rather than increasing it.
Test Your Understanding
Module Assessment
5 questions · Score 70% or higher to complete this module
You can retake the quiz as many times as you need. Your best score is saved.