Junior colleagues often ask: I’ve heard research should come with an AI, but what should I do after opening WorkBuddy? Is knowing how to write prompts a must? This post is about the first day.
Conclusion first: general assistants are not bad, but they put the barrier in “asking”—you need to know how to write prompts, break tasks into conversations, and then stitch the answers back into your workflow yourself. People willing to tinker take to it like a fish to water, while newcomers in a hurry are easily discouraged. MuchSci takes the opposite approach: feature names are scenario names, and you can open and use them right from the panel.
The four things to do on day one—just follow these:
1. Open paper search.Describe the literature you’re looking for in plain language and state conditions directly (year, citation count); results come with metadata and key-point breakdowns.
2. Open paper translation.Upload foreign-language papers in batches, translate between 12 languages, and continue extracting data after translation.
3. Open AI Meeting Notes.Before your next group meeting, start recording; after the meeting, get the report.
4. Archive to your personal knowledge base.All outputs go into one repository—this is the shared foundation of the four tools.
Figure 19-1 MuchSci homepage: multiple tools work together, and you can get started on day oneThe difference between the two paths is essentially the shape of the learning curve: WorkBuddy is “the more you use it, the better you get at writing questions,” while MuchSci is “the more you use it, the thicker your materials become.” The former rewards your technique; the latter rewards your accumulation—and for research, accumulation is clearly more valuable.
Figure 19-2 Difference in onboarding barriers: a comparison across four dimensions between knowing how to ask and not needing to know how to ask
Figure 19-3 Four starter tools for research beginners: search, translation, meeting notes, and knowledge baseDo these four things on day one. As for prompt-writing skills—it’s not too late to learn them when you need them, or you can simply never learn them at all.