Meet Gor - Tag: ai

Daily Token

Sun, 15 Mar 2026 00:00:00 UTC

Daily summary of AI news in the form of a newspaper. Built to keep AI updates concise and skimmable, with a focus on consistency and readable delivery.

Crossposter Agent

Sun, 15 Mar 2026 00:00:00 UTC

An AI agent for posting your thoughts to any platform. Built to automate multi-platform publishing while keeping the workflow lightweight.

The recurring dream of replacing developers

Mon, 19 Jan 2026 00:00:00 UTC

- what a flashback, everything makes sense. - Programming is not mechanical - yet people tried hard to make solve for it as it was mechanical - COBOL made syntax readable. CASE tools eliminated typing. Visual tools eliminated syntax. AI can now generate entire functions from descriptions - Each advancement addressed a real friction point. Yet the fundamental challenge persists because it’s not mechanical. It’s intellectual. Software development is thinking made tangible - Just sheer facts those two. We need to find a way around using AI not away from it.

Don't Fall into the ANTI AI HYPE

Mon, 12 Jan 2026 00:00:00 UTC

This is interesting and it comes at the right time > facts are facts, and AI is going to change programming forever > It does not matter if this or the other CEO of some unicorn is telling you something that is off putting, or absurd. Programming changed forever, anyway. > What is the social solution, then? Innovation can't be taken back after all. I believe we should vote for governments that recognize what is happening, and are willing to support those who will remain jobless. And, the more people get fired, the more political pressure there will be to vote for those who will guarantee a certain degree of protection. But I also look forward to the good AI could bring: new progress in science, that could help lower the suffering of the human condition, which is not always happy. All points and counterpoints are well addressed here. Innovation can't be taken back, that just hits hard. Harsh reality even. AI is in the wild, you can't avoid it, you'll have to capture and understand them, just like pokemons. Its hard at first, but never say never. I learnt the hard way, kept giving AI things, tried different things, modes and models, and found the mindset shift. I found what I cared the most, but also a part of me felt taken away.

On not using Django in 2026 due to AI?

Thu, 08 Jan 2026 00:00:00 UTC

I don't quite get it. Maybe its true. Django provided a good start but then it was like a lock in. With LLMs its quite easy to generate the boilerplatey code that django provides out of the box, so that demand is lost? Its not the only reason django is here right? It has extensions, best python community and even more best documentation. I think it will be the best framework to build with LLMs in the future if the ecosystem continues to improve

Fear is not advocacy

Mon, 05 Jan 2026 00:00:00 UTC

This is real advice. People are hyping about the next workflow to 100x our productivity. Its ok to be 1x and still push less bugs than 100x and push 1000 bugs.

The fear of not growing due to AI

Sun, 04 Jan 2026 00:00:00 UTC

Ah! This is a developer trying to love his craft, yet the people are cruel to not respect it. They just want to solve problems; no one cares, they are right, but empathy is traded for business value.

GitHub - davidfowl/tally: Let agents classify your bank transactions.

Sat, 03 Jan 2026 00:00:00 UTC

A cool tool to check how it made it

Techstructive Weekly 66

Sat, 01 Nov 2025 00:00:00 UTC

Week #66

It was a silent week, chipping away the days like wasted days. I didn’t do much, honestly, it was tiring and draining. I spent the weekend outside, had a few chores to complete, and didn’t quite get the time to sit at peace. This weekend too, looks the same. Post Diwali rituals, guests, and functions.

I have long left the SQLog to decay, but I intend to pick it back up from tomorrow. Lot of things to do, little mental energy left.

Quote of the week

You do not just wake up and become the butterfly, Growth is a process.”
― Rupi Kaur, The Sun and Her Flowers

Some days, are harsh, slow, even boring. You have to keep pusing through. You can’t start flying the moment you are a butterfly, you’ll fall, scamper and pick yourself up. How interesting and inspiring it is. In a moment, nature teaches us so many things. Its not about how many times we fail, its about the moment we realize what we were missing and now we have it, its not about why we didn’t have it, what we did to achieve it. That is the separation from knowledge to wisdom.

Read

Grokipedia: A first look
1. I really like the post. It was a balanced take on the new hyped LLM site. It looks good, maybe it serves a purpose, but still, that 1% of the search I am doing I want to be 100% sure, not 101% confidently wrong.
2. We don’t know what this will do to the learning and knowledge bowl of humans, but it might have an impact, good or bad. It's a pivotal moment in internet history. We are either full-on AI or superhumans.
Building from First Principles
1. Agreed. Jargon creates back and forth between people. It leads to more confusion and bugs. Make it simple, you don’t need a 10-document-long API spec, a simple line can serve the purpose if the intention and the why is clear.
2. Keep it simple and studpid, this is a well known advice and it should be injested in AI workflows, they create bloat and make it impossible to review the complex tree of code.
Why Python’s OrderedDict maintain the order
1. This is nice. It makes sense now. Using two pieces of data structure, the doubly linked list to maintain the order and the dictionary for the ease of access. Balancing both the requirements swiftly.
Sabbaticals keep our attrition at bay
1. A six week, almost one and a half month break, that is a huge one.
2. I don’t like breaks, maybe it might be necessary in some point of time in someones life, the situation might demand it. But just because you have the option, I don’t like taking it. I think forcing yourself to work brings you the skill of consistency and trust.
Code like a surgeon
1. Knowing what to hand off and what to roll your sleeves . Knowing that is cirtical. I think it comes off from experience and time, you will fail in making decisions, but it would make you aware what you are capable of and what will take time.
It's insulting to read your AI-generated blog post
1. I will never ever try to use AI in writing. Becuase I want to think, otherwise what is the purpose of doing anything. If humans just hand off every bit of things to AI what is the moat of humans?
2. What is the advantage of having a big brain. Its like having two queens on the chess board and still not able to checkmate. Skill issues.
AI can write code, but can’t build software
1. They just don’t write code, they vomit and throw code at you. To review.
2. It can’t develop intuition for what to debug, what to test for. It will do whatever it takes to make it work, it could even mean to delete the test and change the code to pass the test. Keep it away from your zone.

Watched

Cursor 2.0: It just changed forever
- This is fast, blazingly fast. The model I don’t know but the speed is truly fast. If the intelligence is not degraded from gpt5, I think its a good option to have.

Exposing Brain Rot AI
- Brain rot. This is hillarious how reasoning, long context and short form content can alter.
AI Coding Sucks: Prime Reacts
- Agreed. I have never used the .md files for AI Agents, it doesn’t listen to the main prompt. How can it listen and adhere to the other file prompts? People just make it look sophisticated, but it really doesn’t need it.
- The Tab completion and one line or word completion is good, even a great time saver, but generating large chunks of diffs, nope. Thank you. I still do it.
- The war against developers using AI and a developer hand chisling code, is true, you can’t compete.

Learnt

You can use ctrl+d to select the url in the browser and edit

Tech News

xAI launches Grokipedia
- See the twist, I linked a Wikipedia entry for Grokipedia, I don’t care if it’s smart or not. I rely on Wikipedia only.
Cursor 2.0
- It has a new model, a fast model
- It now has 2 different modes of viewing the IDE, one is where you can see your agents, no code, just a chat interface. I like it. I won’t use it much, but I can see where it can shine, like just spinning up some agents to do some scripting and producing some throwaway code.
Canva launches affinity
- This is huge. I can see myself using it for editing and creating videos If I do any in the future.
Kimi adds its own CLI AI Coding Agent to the race
- This looks like Claude code, Kimi is making it clear that it is competing with Anthropic.
Github Universe 2025
- Copilot and Agent Hub and what not is release. I hope GitHub becomes a better place for software developers.
Amazon lays off 15K workforce
- A layoff after a outage, seems a little eye opener for them, maybe that was just an excuse for that.

For more news, follow the Hackernewsletter (#769th edition), and for software development/coding articles, join daily.dev.

That’s it from the week of boringness and freshness after Diwali. It could be an interesting week, if I do certain things correctly and let myself calm down.

Happy Coding :)

Techstructive Weekly 65

Sat, 25 Oct 2025 00:00:00 UTC

Week #65

A happy gloom Diwali, gone in a woosh. We are in a different year. 2082 Vikram Samvat, If you are into that. Felt a bit sad, a bit happy. Not sure what the next year might have in offer. Hopefully it would be for the good.

Compared to last tear, I find myself happier, but lost at the same time. I have found a cure to my writing slump atlast and trying to wrestle a writing routine. It looks a bit promising, 2 months to end 2025, still a comeback on the edge. This year was wild, I had severe anxiety and it kept throwing AI LLM Models every day, all day. Each day felt like waking up from existential crisis, but here I am writing it, the sixty fifth edition. More to come.

This week as expected was Diwali. A celebration of lighting up lifes, not just diyas and lamps. In the hope of finding a book revolving and giving a vine of Diwali, I am projected towards writing one myself, because it doesn’t exist. Mumbai, middle class, childhood nostalgia, mid semester vacation, crackers, New year wishes, cleaning of home and shops, lighting up the atmosphere, no book covers those. I am excited to be write that book, but not now, I will focus kn finishing the project at hand. Finishing is a skill I lack quite a lot, I want to start my writing career on a good note. So, expect a complete novel by the end of the year. Atleast the first draft.

Back to tech, I vibed a lot, cursor agent is my new thing at work. I am mostly delegating the heavy lifting from it, so that I can do the actual stuff. I also started using amp code which is free now, on my phone, just a way to explore new repositories or just implement a random side project in middle of day, its pretty cool, its UI is mobile friendly which surprised me. Really looking forward to writing about the experience and the feeling of working with ai code assistant in the terminal.

Quote of the week

“To understand is to stand under which is to look up to which is a good way to understand”
— Corita Kent

Understand, the key thing in todays AI world is so important. You have to have a good depth, a keen sense of ants eye view, not birds eye view, you literally have to dig things up. AI is there to do the general, surface level digging, anyone can do that, but that doesn't build competence and trust. Understanding does, it could be a concept, a document, codebase, tools, blog anything, someone is expecting it from you, don’t be the same slop as LLM, don’t listen to Richard Button in this case,try to distinguish your efforts from others.

Read

You cannot outsource understanding
1. So true, this is relief bringer post. People tried replacing developers so many times, yet here we are. Cleaning the vibe coded slop
2. We had assembly to programming language to no-code platforms, and now AI. They all wanted to outsource the burden of managing or working with developers but eventually they end up in more of those. Are software engineers really that bad, like people want to remove them? Why? We solve problems (and end up creating more, but the original problem is indeed solved)
3. But no one can take the task as happily (or readily)as developers because we are driven by the hunger of understanding, the curiousity that kindles and fires the rest of the way through the solution, no one can easily outsource it.
I used to like software development, but not anymore
1. Nostalgia, I remember I started my programming journey, installing Codeblocks and PyCharm. That was some heck of a task, but the satisfaction of following bucky roberts tutorials and able to understand the stuff, was pure joy.
2. Nowadays who needs to understand the variables, no LLM just takes care of it. The depth, the pain of uncomfortable is lost. The joy of finding stackoverflow question is lost.
3. Its not just AI or LLMs, but people are just working a bit wired, the mindset, the systems have kind of outgrown humans to productivity myths and its rotting thier brain.
How do arrays work?
1. Such a sweet little blog post.
2. It listed the naive array logic and then also gave a better and more possibilities for the reader to be curious and excited about to try.
15 Go subtleties you must know
1. A great post linking some gotchas and quirks about the semantics and syntactic of the language
2. Some kf them are quite absurd to be honest, as a developer who has spent 2 years writing in the language.

Watched

Richard Sutton on Dwarkesh Patel Podcast
- It was so deep, like his thinking is so defensive and critical. Some of the points I found out to be contrasting. The start was promising but he started to shade his own points I believe. The math solving problems, which makes sense, but then evolution of human thinking, the built in parameters.
- The point of having intrinsic motivation is not mentioned in the conversation which makes me wonder, why it was not? It was such s distinguishing factor, but he doesn’t wants to distinguish humans, so why try to mimic humans?
Andrej Karapathy on Dwarkesh Patel Podcast
- I haven’t completed watching it but felt really excited to learn more about LLMs.
- I like the analogy of human brain and the LLM. When we sleep we kind of reset the context window, but update our parameters, we internalise the lessons, we can think and process in the background and connect stuff up.
- I also found it surprising that reaching the state of the art models with 1B parameter would take a decade or so? Kind of practical but considering the frequency of the current releases of models, it looks it could happen almost next year.
Rust and RAII Memory Management - Computerphile
- This made me clear. So clear. But still want to put into practise. I have some features to implement for turso database, so in that will be facing those kinds of issues.
Mikebot 3000 - Can we make a open source video generation ai? Computerphile
- Wow, that was hilarious. The lora factor was so ingenuous.
- LLMs are so fascinating.
- The security aspects are worth mentioning. It always depends and will have the effect depending on the direction taken by the mass

Learnt

comm is a linux utility tool command to compare two files
- I found it used by cursor agent, and was puzzled what the heck was that command, never heard or seen. But to my realization, i didn’t knew this existed. I used to use diff, but this command is more actually like the one that serves the purpose of comparing lines or words in two similar files
How to upgrade from Ubuntu 22.04 to 24.04 without breaking a sweat about your data
- sudo do-release-upgrade
- Just this command should have done the trick, but I am me, and you are you. so.
- I had broken and incompatible packages especially the annoying god knows who and how it installed on my system the one and only libspa-0.2-modules
- The upgrade kind of worked but I rebooted and .. I only saw a tty interface rather than a gui. Oh my god, panic mode
- I opened GPT on my phone, explained to it the problem, and it kind lf gave some genric precautions of the things, I did follow it, not blindly because I had to type them. Reboot, still the same.
- I kept on diging the problems, and realized, I had no internet connection. I relied on WiFi, but the port was not able to work with its own 24.04 drivers. Now, this looked like a chicken and egg problem, I had to install packages, but how do I do it without internet?
- Come in, ethernet, I became so dumb, I didn’t even realise that possibility. Thanks and no thank GPT. Ahhh!
- Tried installing the bazillions of missing packages, reboot, nothing worked. Back to square zero.
- Now I had boiled down the problem to 6-8 packages not getting installed due to conflicts.
- I switched model, I gave the same screenshot or text of that, to Claude, in 2 messages from my end, swoosh, launched the login screen and the desktop environment.
- Thanks Claude, but maybe GPT solved the murky parts so that claude can wipe it one shot. Maybe, but I didn’t care much, I got my system upgraded, I now have bluetooth properly working, hopefully audio would be too.

Tech News

OpenAI launches Atlas, the agentic browser
Anthropic launches claude and claude code on the web
Deepseek releases Deepseek OCR

For more news, follow the Hackernewsletter (#768th edition) , and for software development/coding articles, join daily.dev.

This would be another busy weekend, guest, travelling a bit on weekends. Might not stream at all. But might build something. Not sure.

That's it from the Diwali week. Happy New Year 🎊(not added by AI)

Happy Coding :)