It’s been a little 2 months since my rant about shipping nothing in the age of AI. This blog is going to be a self-evaluation piece along with a few things that I learned along the way. I still find myself going back to my previous blog because it serves as a north star in the labyrinth of living in 2026.

Overtime I realized that writing it all down only and publishing it off as a blog did not auto-magically clear out all the distractions for me and made me a 100x execution machine. What it did in fact is put me on a path of 1% improvements, if and when I find them.

Early realizations

1. Overcoming the doomscrolling tendencies whenever watching a youtube video on a mobile phone isn’t easy. There are singular instances every other week when I tend to spiral out and lay motionless on the bed just scrolling through in hopes that I find meaningful content.

2. I have realized that I am craving for content that shares a new perspective, is attractive and eye catching, while retaining intellectual value. In other words, I am hunting for the top 1% of content.

3. Somewhere along the lines, my understanding is that one does not have to participate in the hunt to access such valuable content. Reading books, blogs and having conversations with chatgpt that is tailored to give me quality answers far outweighs the scrolling tendencies.

4. Keeping your thumb 👍 inactive is ironically the path of higher resistance! So now, I am adopting books and creating this alternate circuitry to tire my brain out with meaningful thoughts rather than over-stimulation.

Success rate ? Probably 70% over 2 months but it is a start.

5. I am grateful about the fact that I do not hold accounts in Instagram, Snapchat and open LinkedIn around once in a month or only during hiring season.

6. I realized that I now had a Twitter/X problem. Having been a twitter user so that I can consume textual micro-blogs instead of reels has been my line of defense against short-form content in other platforms. Over the period of past 2.5+ years, I have carefully curated a feed filled with software engineers catering to different domains like UI, graphics, databases, cloud and backend.

7. However, since late 2025 and throughout 2026, the signal to noise ratio on twitter has fundamentally shifted with more and more people producing content around every new thing that the AI world has to offer and this became my Achillies Heel. The only way to combat this was to actively unfollow a lot of accounts and popular creators and re-gardening my feed.

This began my ROMO - relief of missing out!

8. I started finding joy in the fact that I am not constantly bombarded by AI news and slop. Now, I have come to terms with the fact that I do not mind trying out a shiny new feature or a product 3-4 months down the line of it’s launch. Giving it some soaking period and allowing people to form opinions on it and then using that as a baseline feels better and more productive instead of running on fumes and being on the bleeding edge of everything.

Things that changed how I think

1. I take notes. I think somewhere earlier this year, I had stopped taking notes of things and that meant I was experiencing things in a shallow manner. Interestingly, I also realized that note taking allowed me three things

   1. My right hand is holding the pen and not sitting on the arrow keys to fast-forward the long form youtube video that I am watching.
   2. Taking notes means naturally slowing things down to your writing speed and taking pauses. This means that I am consuming every chunk of content meaningfully and there are natural barriers to over-consumption.
   3. My brain isn’t being coaxed into running at overclock speeds to consume everything within a short time. Operating at less than max capacity means I can operate for longer hours while respecting my cognitive bandwidth. To put this into perspective, accept that your brain is not an F1 car, the incentive is in navigating roads and not spinning around race tracks.

2. Prompting AI but also scribbling. I think we are deluded into thinking that we need to review code or plans at the speed of token generation. Pre-LLMs era, humans would tend to get competitive over who can get how much work done and this competitiveness has become a part of our DNA. However, this is becoming a problem now because we are speed-reading the results of a prompt. This is not productivity!

Having a notebook beside and scribbling down some of the core ideas/plans, or drawing boxes and flowcharts is helping me better process things rather than naively outpacing a token-generator.

When it comes to code, I try to make sure that the diffs that I am having to review are preferably under 500 lines by actively prompting it to break down tasks. As a developer, I rarely appreciate massive PRs/changes by my peers; why should my LLM be treated any differently ?

3. Avoiding skimming. Throughout our academic lives, we have been told to quickly skim through things. Be it passages on a text, the entire question paper on a exam or code documentation. I strongly feel that one needs to unlearn some parts of it. Consuming deeply and meaningfully is important. You do not have to go through 10 videos and learning everything in a day, do 1-2 properly.

Things that changed my environment

1. Removing social media during work. I have actively kept WhatsApp in a different browser profile which I try not to open during work unless absolutely urgent. My “work profile” only has meaningful tools that help me to get my job done. Apart from this, I also maintain a “study profile” which I use for learning things on the after-hours of my job. Additionally, I keep the phone either face down or behind my laptop so that I do not see it thereby avoiding the distraction.

2. No multitasking. Humans were never designed to multi-task. We just context switch at lightning speeds. In 2026, I am giving a lot more time to my writing than I previously did. This is manifesting itself in the form of various blogs. This act sets the precedent of what I can do if I work deeply because every blog takes about 4-5 hours of thorough attention followed by proof-reading before publishing. I have noticed that the unlocking of this result has started to percolate into my other activities thereby diminishing low-effort and half-assed work.

3. White noise as an antidote to mind wandering. Once you learn to sit down and work for 1-2 hours at a stretch, you start realizing that your mind tends to start wandering off despite the lack of stimuli. This can be attributed to the fact that the brain focuses and de-focuses. It’s a natural peak and dip cycle that everyone goes through. We cannot shut it down but we can elongate the time interval between a peak and a dip. That’s what white noise has helped me with.

That’s all from me today. Take care.
Ritesh Koushik