I used to use Feedly for a long time to keep up to date with my favorite blogs. Six months back, one fine day, I started working on my own RSS feed reader in my free time. It went somewhere but my reading time went down. Sadly my reader didn’t work out - it has hindered my reading and now I am out here resuming my reading.
I have mashed up a new blog reading system for me but not based on my reader. Maybe I will talk about the new system some other time after experimenting with it for a bit.
But for now, here is what I read from 27th to 31st July 2026.
Blogs
-
I am on the internet for gold mines like this.
-
There is a DOCX to DOCX converter, wait what?
-
I have come across RACI/ARCI & DRI at work before. I still feel like it is “too much theory” but the author gets to the practicality of it. “Assign one accountable owner” is a distillation that I like to remember from this post.
-
Love multiple takes in this one. I do find myself in a similar position to the author on the topic of taking longer times to read AI written artifacts. There was this particular line that caught my attention:
Using AI to make your work easier while making other people’s job harder
My take: As we use AI more and more in a team setting, we need to be very thoughtful and mindful in producing artifacts that don’t overload our colleagues.
Mastodon, The Only Good Choice
Rekindled my interest in Mastodon. While doing my re-exploration of Mastodon, I met https://indieweb.org which made me double down on starting to write a blog post like the one you are reading now.
Generated and suppressed demand
I suggest reading this post first - something that I have read in the past that gave me a good mental model of team performance. I haven’t fully internalised the concepts in this new post though.
-
Beautiful short memoir. Reminds me of what is important.
Product for Engineers is now build mode
Long time reader of this blog, so read this out of curiosity to know why and what they are doing by renaming their blog.
-
The most important skill in prompting is expertise in the domain you’re prompting for
Don’t follow what LLMs tell you. Ask critical questions that improve your understanding of the codebase/system. That way you can steer the LLM in a much better way.
-
My rate of sending cold emails is very low. But there was one time where a response to my cold email completely changed my world view. Maybe I need to try to send more of these?
Oh, I have also gotten a few cold emails from people around the web and have genuinely enjoyed those conversations.
-
My fav parts were 4 & 5. Because those are the ones that I have been coming to realise more and more in the recent times.
The best strategy is to work on what you like, and what you find interesting and energizing, because curiosity, enthusiasm, and craft compounds
What should a personal website be?
Spoiler: It has to be whatever you want it to be.
-
Great article, especially relevant in the age of AI where software teams are expected to move with “speed”.
Books
Rework
(read 82 pages)
I haven’t been taking notes for this one. Because it is the kind of a book that you could re-read instead of reading your notes.
Let Go
(read 38 pages)
My wife picked up this book from the book store. I randomly started reading on a weekday night. I could visibly feel my “stress level” going down page after page. wow, books are weird - they can alter space, time, and mind.