pravj

Recently, I visited Budapest and read a book with the same name by Victor Sebestyen.

One critical topic stood out: "Budapest, the city, was rebuilt significantly in the two decades leading up to 1896 for a World Exhibition marking 1000 years of the country."

It was not a proper world fair, but we can call it so.

Since then, the same question has been running through my mind: what happened to world fairs, and why don't we have more of these?

Operationally, it gives me chills to imagine a "concert/festival" [in today's terminology] of such scale being pulled off across months.

## Tab Highway: Browser extension focusing on the UX/process of decluttering

It turns tab cleanup into a game.

When you pile up too many open tabs or whenever you want, it opens a night drive where each tab is a road sign you drive through or send to the trash.

[GitHub](https://github.com/pravj/TAB-HIGHWAY)

This is still a work in progress. It works on my machine, but not fully distribution-ready.

## A short satirical blog on the overly-expressive/interactive web experiments fueled by coding agents.

Ironically, the project listed above falls in the same category.

[Woodworking Has Been Achieved Internally](https://hackpravj.com/blog/woodworking-has-been-achieved-int...)

I used to envy and take inspiration from other workplaces where good [but not necessarily good-only] writing was respected; where a pre-read is really read before the meetings, thoughtful comments were made on it, etc.

AI workflows have obviously simplified documentation generation along with the code, but we had to work on our product/engineering practices to generate meaningful documentation, and not just vestigial/temporary documents in the process. On this particular point, we've made positive progress lately.

https://postimg.cc/MXBQqrXf

One main takeaway from the book was that "you can just look at the ratio of turn-taking duration, and which speaker/participant is spending how much time" to decide "what happened in the conversation".

The same goes for AI-generated conversations; verbose responses are the default behavior, and models are incentivised to keep that output-token ratio. Too easy to catch/notice, pretty annoying.

PS - I work in the Conversational AI space, and it is quite an effort to keep the ratio right for people to stay long enough on the phone with AI agents.

The popularity ranking matches the quality of content produced, and people are spending more time than anticipated on Grok and Roll to confirm if they (listeners) are hallucinating or if the radio is really stuck on roll.

Recently shipped this personal art project that turns daily Wordle attempts into gritty / struggle-filled stories, kinda similar to the emotional arc of the Wordle game play.

You can upload your own Wordle game screenshot to generate one for yourself.

In addition to completing what was once in the idea list, I got to learn about

- Prompt fine-tuning: Models are sharp enough to complete Wordle games quicker than human average scores, so I had to dump that down and get the average down.

- Karpathy’s Autoresearch: Experimented with auto-research for prompt fine-tuning, in addition to manual prompts.

- Vision models: While leading labs have multimodal models with quality visual reasoning, the benchmarks are still quite different for a simple Wordle analysis (reading what letters were yellow/gray/green); I also noticed labs/companies with separate vision models but their APIs lag significantly compared to what’s possible in developer experience.

- Video generation: For the last few days, I have been experimenting with automated video generation for the project's social handles. I'm still struggling with the right hooks that reduce the skip rates, but it's fun.

---

Additionally, working on an Apple Watch app similar to my Mac app on the same lines, [Plug That In](https://plugthat.in), i.e., notify before the device goes too low on battery, but with a twist.

From my failed attempt, I remember that

- Students had to find a project matching their interests/skills and start contributing early.

- We used to talk about staying away from some projects with a low supply of students applying (or lurking in the GitHub/BitBucket issues) because of the complexity required for the projects.

Both of these acted as a creative filter for projects and landed them good students/contributors, but it completely goes away with AI being able to do that at scale.

Using this to maintain my writings, projects, readings, and curated photo collections.

Not going to deny that it was a bummer to read your second line, lost all the Endorphins earned from reading the first one. Nevertheless, I want to fix this if it's a legit issue.

Do you mind sharing more details on the battery consumption from the app?

I'm using IOKit power source notifications and hearing this feedback for the first time. (personally using this on my work [old M2] and personal [new M4] MacBooks for the last ~2 months).

An annoying little laptop charging reminder utility that does the job.

---

There are times when I'm deeply immersed in focused work, a meeting, or engaging video content and end up missing the usual low-battery notifications on my MacBook.

When the laptop suddenly shuts down, it's followed by the familiar and frustrating walk to find a charger or power outlet. It can be annoying and occasionally embarrassing, especially when rejoining a session a few minutes later with, "Sorry, my battery died."

Over the past few weekends, I built Plug-That-In, an app that introduces "floating/moving notifications". These alerts follow the cursor, providing a stronger, harder-to-miss nudge regardless of what’s happening on screen.

The app also includes a few critical features:

- Reminder Mode: When the battery reaches critical levels, the app emits a configurable alert similar to a car's seatbelt warning, continuing until the battery is addressed.

- Do Not Disturb Settings: Customize alerts and sounds based on context, such as when system audio is playing, a video is active, or the camera is in use.

It grew out of a personal need, and I'm glad to see it used by over 50 people in the past month.

An annoying little laptop charging reminder utility that does the job.

---

There are times when I am deeply involved in a focus-work session, a meeting, OR watching some sort of engaging video content, and don't pay timely attention to the standard low battery notifications from my MacBook.

After the laptop shuts down suddenly, what follows is the most annoying walk to find the charger or the charging outlet. It's frustrating at times, sometimes embarrassing because you have to say, "Sorry, my battery died down" as you join back the session after 2-3 minutes.

Over the last 3-4 weekends, I have been building Plug-That-In, which has floating notifications. Essentially, a notification that follows my cursor movement, so I get a stronger nudge irrespective of what I am doing.

There are a few other critical features, such as Reminder Mode and Do-Not-Disturb Settings.

- Reminder Mode: On critical/lower battery levels, it will keep beeping like a car's seat belt alert for some time (configurable) when the battery is really low.

- Do-Not-Disturb settings: Configure what sort of alert/sound it will generate when I have system audio playing or video playing, or the camera is active.

It has addressed a personal need and has already proven useful a few times over the last weeks.

Standard system notification comes at about 10%, and most of the time, in my case at least, whenever I miss that, the result is "laptop shutdown amidst an ongoing video meeting" or something like that. (Basically, too late before I act)

Just so that I don't miss the reminders, the app will show an overlay window with some text, following my cursor, and a custom sound.

I built a version this weekend, and am current doing a dogfooding exercise.

Overall, people are making a net-negative contribution by not having a sense of when to review/filter the responses generated by AI tools, because either (i) someone else is required to make that additional effort, or (ii) the problem is not solved properly.

This sounds similar to a few patterns I noted

- The average length of documents and emails has increased.

- Not alarmingly so, but people have started writing Slack/Teams responses with LLMs. (and it’s not just to fix the grammar.)

- Many discussions and brainstorms now start with a meeting summary or transcript, which often goes through multiple rounds of information loss as it’s summarized and re-expanded by different stakeholders. [arXiv:2509.04438, arXiv:2401.16475]

Users in a white-collar setup are limiting their relationship with LLMs at a peer-level, and a multitude of AI-tool UX is designed to reinforce that same hierarchy, which is marking the entry of "AI Slop" into the workplace.

I stopped writing right after college when I moved to full-time work, but I want to change that now.

Just as I started, I realized the website was hosted on Forestry CMS, which has been discontinued, so I will have to figure out how to maintain the URL structure, etc. (this is important since there are a few popular/top-10 pages for certain queries/searches [1]).

[1]: https://hackpravj.com/blog/solving-semantris-opencv-word2vec... (for "Google Semantris")

- 50 jumping jacks [3 sets]

- 25 knee-highs [3 sets]

- 30 mountain climbers [2 sets]

- 25 pushups

- 1 minute plank

I stopped doing it after I was able to meet my weight loss milestone, need to restart.

Do you have any learnings to make such initiatives work?

All of them are for January-February-March 2021. Does that cover the time-bound aspect you're talking about?

- I assumed everyone here on HN would be familiar with him. But lesson learned that it was just another assumption. I've added the details about him.

- I've fixed the broken links.

- that lets you (as a startup) optimize spending on day-to-day tools (similar to optimizing your cloud spendings)

[1] https://status.twitterstat.us/

[1] https://quickdraw.withgoogle.com/

I like to call it a fundamental or building block technology that enables you to develop things on top of it, my favorite being a finite-state-machine we modeled in a sheet. Won't be surprised if we end up talking about Google sheets being one of my favorite products.

Can you explain more on the following point, in the non-personal information section of the policy?

"navigation paths used while visiting the website or platform"

Vernacular.ai helps enterprises streamline their contact center operations by deploying multilingual voice bots and delivering up to 70% cost cuttings. Our product, VIVA (Vernacular Intelligent Voice Assistant) uses cutting edge speech recognition and natural language processing technologies allowing it to understand user's queries and resolving them through engaging conversations.

- ML Research Engineer (https://vernacular.ai/career.html#jd-mlre)

- Associate Product Manager (https://vernacular.ai/career.html#jd-apm)

You can apply via email as well, [recruitment at vernacular dot ai].

Reach out to me, [pravj at vernacular dot ai] for any further clarifications.

I am planning to make it more accessible, please follow the follow-github-organisation[1] for updates. Pun intended.

Let me know if you have any other feedback.

[1] https://github.com/follow-github-organisation

- Using GitHub webhook, an event is triggered to see if the org is already being followed

- If not, a record is created (organisation name, public repo count, issue id etc)

- Every 6 hour, (for all organisations) a cron job checks if the org has any extra public repositories

- In case of any new repository, a comment is created with the links to the new repositories