AI-Productivity-Speed-Run-with-Dave-Katague
The slide deck presents a practical framework for boosting personal productivity with large language models (LLMs). It begins by highlighting three critical transfer points—Input, Process, Output—that govern every human‑LLM interaction and identifies common bottlenecks such as slow manual input, disorganized documents, and reliance on outdated tools. The deck then classifies LLM operations into Reductive (summarizing, extracting), Transformative (translating, tone‑shifting), and Generative (brainstorming, outlining). Building on this, it offers concrete tactics and a curated toolbox to reduce input friction (voice transcription, OCR, conversation exporters), improve the processing stage (voice‑enabled agents, prompt‑building utilities), and enhance recall (clipboard managers, emoji tagging, text‑replacement apps). A bonus section demonstrates how to use Google AI Studio to let an LLM “watch” video or screen recordings and perform deep content analysis. The presentation concludes with actionable next steps—download the slides, question existing workflows, and experiment with new tools weekly—while inviting the audience to follow the author for ongoing tips.
Topics
Input‑Process‑Output Framework & Human Bottlenecks
Introduces the three‑stage model (Input, Process, Output) for any human‑LLM interaction and pinpoints the main friction points: manual typing, scattered documents, and legacy tools.
LLM Operation Taxonomy: Reductive, Transformative, Generative
Defines three core operation types for LLM conversations—Reductive (summarizing, extracting), Transformative (re‑formatting, translating), Generative (ideation, outlining).
Toolbox for Reducing Input Friction
Curated list of transcription, OCR, and conversation‑export tools (Wispr Flow, VoiceNotes, Granola, Text Sniper, Poe/Claude/ChatGPT exporters, iTubeGo) aimed at speeding up the way users feed prompts to LLMs.
Enhancing the Process Stage with Conversational Agents
Recommendations for voice‑enabled agents (Grok Voice Mode, ChatGPT Voice Mode), Chrome extensions (VoiceWave), and prompt‑building utilities (Distillation Prompt Builder, Whisper transcription) to improve how LLMs interpret and transform data.
Boosting Recall – Output Management Tools
Techniques and apps for organizing LLM outputs: emoji‑tagged titles, cross‑device clipboard managers (Paste, Pasteboard), text‑replacement utilities (Rocket Typist, Espanso) to make retrieved insights instantly searchable.
Bonus: Turning LLMs into Video‑Analysis Engines with Google AI Studio
Step‑by‑step guide to upload or link a video in Google AI Studio, break it down token‑by‑token, and iteratively feed insights to other LLMs (Poe, Claude) for deep analysis of visual content.
Actionable Next Steps & Community Invitation
Calls the audience to download the slides, audit personal bottlenecks, question existing habits, test new tools weekly, and follow the author on social platforms for ongoing experiments.