Learn Anything Faster: Save 80% of Your Time with Gemini Deep Research & Obsidian [Guide & Setup]

okay this is an Obsidium Knowledge Graph  automatically created by AI and the map is  

not the only thing you get we also have a clear  learning road map clearly explained section by  

section with key Concepts and various articles  and their reference links as well addition to  

that you will also have an index page that for  you to complete the notes of each key Concepts  

for me personally I cannot think of any other  ways simpler than this for you to learn a new  

topic in this video let me show you the  simple setups that could help you get the  

similar learning road map within 10 minutes

Hi friends my name is Tony. I'm an a engineer and tech nerd enthusiastic in productivity hacks

I guess you may have seen prompt tricks and tips to ask GPT to teach you something

new but how about serious topic research searing and structured learning process with  no applications like obsidian recently I found  

that combining Google Gemini deep research and  obsidian you can save loads of time to teach  

yourself any topic today's sharing will include  two major parts that can help you learn anything  faster

topic research with Google Gemini  building obsidian Knowledge Graph with AI in cursor

okay if you want to learn anything  from scratch the first thing you will do is  

research relevant learning materials from online  then trying to piece together various fragmented  

information to create your study guide this will  take you days or even weeks of effort to research  

but what if I'm telling you that this can be done  in minutes with Google Deep Research

so what is deep research? Deep research is a function  that supported in Google Gemini application  

it is hidden below here okay let me explain  in a slide so traditionally when we um search  

uh anything online and search engine like Google  will help us search relevant website or if you use  

perplexity an AI search engine it will look for  website and read relevant materials and picked up  

for you but that is only the start so imagine um  that you have a smart research assistant that they  

do not only use Google or Perplexity to complete  their research they also plan how to do the  

job better so here comes the Deep research it is  an AI agent in information searching so as a user  

you will ask a question say what is artificial  intelligence then Google Gemini will create a  

plan first and you as the master will review  the plan to see whether it is good for you or not

If you think the plan is good enough it will  follow the research plan search online read the  

materials and write a report for you to review so  let me show you how to do it say I would like to  

know everything about the Starship you can simply  ask "I would like to know about Starship of spaceX..."

okay so currently the AI behind this is  thinking of a research plan could help you  

better so oh so we find information about the  SpaceX Starship program including its goals and  

current status so this is the first research  Target then they will find articles and news  

report on recent Starship development including  test flights and upcoming Mission then finding  

information of the technical specs of Starship  the size the capability and find information  

about potential applications as well and last  but not the least the environmental impacts  

the potential challenges and risks associated  with Starship as well so the AI create a plan  

for us if you don't like this plan you may be  edit it as well

so I say I will like to edit the plan

about so I would like to know about the  Elon Mask's Vision on Starship as well  

all right so we have an updated the research  plan so at the I think at the last re  

at the last action point it added fin  information about Elon Mask's vision on  

Starship the uh the six task above is the  same all right okay now we can click the

start so while the Google Gemini is doing deep  research for us it will search online read all  

the relevant articles and analyze the content  then we'll get back to you a full report for  

you to review so seeing it's researching for  49 pages okay 60 Pages or even more so if you  

want to learn any area, any task, any topic you are  interested in this is the perfect tool is that I  

will let you know when the resch sear is finished  you can leave this chart in the meantime so okay  

so let's have a quick look at the example I will  use later so if you want to create a learning map  

from Google Deep Research this is the prompt I  use if I want to learn what is large language  

model and how can I use large language model be  AI explication would you please create a learning  

road map for me okay so under this prompt Gemini's  task is not only create a research report but also  

to write a learning road map for you based on  the research result so at list a research plan  

for me then write this learning road map. It is called The Learning road map from Basics to  

building AI applications it was separate into  sections introduction to large language model  

there are key Concepts you should know and  the articles that it found relevant and you  

can definitely click the Articles including  YouTube videos as well you see the videos  

introduction to large language model Google  cloud and YouTube let's click this see hello  

and welcome to introduction to large language  mod great all right so if you don't want to read  

articles you can definitely check those videos  later let's check whether our Starship report is

completed okay it's really close

great it has been completed okay let's  check the report SpaceX Starship reaching  

for the stars and Mars with Elon musk's Vision wow  the title is attempting all right say Starship  

development and test fly um I will leave uh the  link to this article below and you can definitely  

check this out so let's check how many sections  it have Starship development and test flight  

recent development uh flight 5 6 7 and upcoming  missions as well so take an example um SpaceX are  

outlined several key missions for Starship: Orbit  Refueling Demonstration wow Uncruited Lunar Demo  

wow W this is exciting too all right and you will  have uh those specs of Starship and uh internal  

designs structure features a lot lot of detailed  information all of this information was actually  

not make up from um Google's Gemini. it was from  reliable sources you can take a quick look at this  

they are Wikipedias they have informations from  Starship official website they also use uh Reddit  

as information Resource as well and of course  we got YouTube videos and NASA report so I I  

would say um if I when I was a student if I have  a tool like this I would definitely have higher  

marks of my assignment because this is amazing  you have those multimedia resources from different  

uh information source and combine together to  create you this single report in just few minutes  

additionally if you have Google uh drive you can  open in uh docs and you can edit you can export 

the report as well uh since obsidian use Markdown  format we can you can choose download with

markdown okay it is downloaded let's open it and  we have markdown file okay I have already import  

the learning road map as a markdown file to  the obsidian as you can see that we already  

have a well researched learning road map but  having a road map is only the beginning that's  

why we need to use obsidian to help us to learn  so the major reason why I love obsidian is that  

you can write your note cards each time by  focusing on one concept then you can easily  

find and build connections between your new nodes  and existing one with time you will have your own  

personalized Knowledge Graph like this on any  topic you are interested in this idea was made  

Popular by book "How to Take Smart Notes": Learning by Writing there are loads of great videos on  

YouTube right now talking about this so I will  not explain that in details but focusing on one  

paino SLOW yes it works but it will take you a lot  of time to build it up and sometimes when you are  

facing a sometimes when you are facing a strange  topic or an area you will feel very uncomfortable  

but you just don't know where to start the best  way is to buying a textbook in our case what we  

need to do is to extract those knowledge point  from the learning road map and make each one  

into note cards for us to write learning notes  now this process can be fully automated by AI  

okay now we use cursor to open up obsidian file  folders Regarding why and how we use use cursor  

to operate obsidian files you can refer to the  link below you can see that this is the markdown  

file from obsidian and uh for each section we  will have an Index Note introduce the concept  

and provide the links to the note card we need  to fill but how can we create it automatically  

this is the part we need to involve AI all right  so the process is very simple it only involves  

two steps the first step is to create index  notes and the second step is to create a back  

link from index noes to the road map what I will  do here is that I will clean I will delete all the

files and remove those back links okay uh now now  we only have one road map nodes um in my obsidian  

folder let's use my prompt to try again this is  a learning road map report I need you to do the  

following step one based on the learning road map  create a list of markdown files for each topic for  

each topic file write an introduction and a list  of knowledge Points each knowledge points title  

should be written in double square bracket  uh you can simply use cloud Sonet 3.5 then

submit okay now we have an AI  is creating each note files for

you okay we have the ethical part the introduction  part this is the file created by AI okay it has  

been completed we already have those uh notes  created with introduction and uh note card link  

listed inside uh what you need to do is simply  accept all right you already have those files  

the next step is to add a back link on the road  map file so we can use it conveniently this is  

the prompt add your created file name list  in road map file place them under each topic  

related remember use double pret to quote  the file name okay submit let's see how it  

works okay it seems that it got run to create  new file for me no worries let's update the

prompt all right let's submit  again so in this case you will  

click continue and revert so it will undo  the previous task AI agent did and try

again all right it is working great it even add an emoji  here related detail notes cool good job so what  

you do accept it and everything is done so this  can be down in couple of minutes let's track the  

obsidian file see how it goes in obsidian this  is the original node llm learning road map and  

uh here's the link edit we can jump to the index  note where uh it introduced the concept here and  

have all those key concept knowledge for you to  fulfill we can try our graph view as well so this  

graph view only uses notes in my notes folder  so what I do I will move those index notes into  

notes see how it goes great we have our road  map now if you want to learn any New Concept  

about the large language model you could have a  quick look around that if you have read something  

about testing autom you could simply click it  and write your learning notes Here step by step  

with a clear road map you can learn anything  faster okay this is today's sharing so we do  

an AI deep research with Google Gemini then  we use cursor AI to build obsidian Knowledge  

Graph thank you guys for your time if you like my  video please hit the Subscribe button and click  

the thumbs up if you don't like tell me how to  improve that thank you very much for the time