abhishekbasu

I highly doubt this skill adds much. Even prior to Opus 5.5 these models were good enough to generate videos, look at the demos on https://fretta.ai for example.

Adding audio and sfx is easy too if you add ElevenLabs to the mix.

agreed, this is a huge pet peeve of mine. every app that has a frontend should have screenshots on their github

Thanks for sharing. Do you have a few examples of these human bottlenecks where a junior engineer is apt?

Interesting read, as someone who chose not to pursue an MBA after my technical masters in '18 I wonder if there is any value aside from time to explore when pursuing an MBA in '26

Thank you for your kind words! Your marketplace idea definitely sounds interesting to me, if not for payments, it would definitely provide a way to improve visibility through easier data exchange due to a common platform.

Hosted on premise for enterprise, VPC for mid market customers. For enterprise customers, FDE model makes most sense for a product like this and they would assist with implementation and training. For mid market customers, I'm still exploring options to make it cost effective.

For domain-specific optimization, the value is in the solver integrations, specific constraints that form the seed, and the modular simulation that powers the visuals. The software is monetized, not the services around it.

> How will you stop LLMs from recreating it? Having worked this space for a while now I think there are two ways to ensure reliability (the real moat here), first is going deep into five-six problems that are complex enough that out of box solutions/simple prompting don't work well. Second, tightly coupling a simulator to provide rapid feedback that actually helps change manage and solve the "people" problem when optimizing operations.

Also, this was a quick fun project for a multi-agent flow https://flickfeast.party/

Developing LogiModel AI (https://www.logimodel.com/) which is an agentic supply chain optimization engine that forecasts demand, optimizes operations, and simulates scenarios to reduce costs while keeping your network reliable and customers satisfied. It integrates seamlessly with your ERP and document repositories to learn your business context, then acts as an intelligent decision engine, freeing you to focus on strategy while it handles execution.

Always interested in possibilities of LLMs interfacing with MIP solvers.

you're right, it should be (3,0) with optimal obj value of 6.

this was a great read to start the new year! having worked extensively with mixed integer programs, it is always a bit disheartening to see them not used enough for everyday decision-making. one of my goals this year is to create a layer to make it easier to formulate mips and test them, via plain text input. this would hopefully increase adoption through a lower barrier to entry.

Great product and congratulations on the launch. Who is the target user vs customer? On the surface, and I may be wrong here, this feels like a LLM layered on top of a typical AutoML structure eg: TPOT, Caret. Is that the correct mental model for a tool like this? And if so, do you see a similar problem that these tools faced in broader adoption at companies?

Grokipedia at home: https://github.com/abhishekbasu/localwiki

I've always had the impression that Mathematical programming esp. Mixed integer programming/Integer programming is largely "unknown" outside of core engineering and operations research. It's an excellent framework to solve a whole host of problems that arise in business and elsewhere, which are solved using suboptimal (hah) heuristics instead.

Okay, maybe I was a bit harsh, but it definitely doesn't pop up as often as deep learning and statistical machine learning. For those who wish to get deeper into this, I highly recommend Optimization over Integers by Bertsimas and Weismantel.

Interesting. Deepseek R1:

You prefer elegant, high-level solutions that are intuitive and accessible to other developers. You likely favor functional programming, clear abstractions, and code that reads like prose.

Abstract ↔ Concrete: +4 Abstract Human ↔ Computer Friendly: +9 Human-Friendly

+1 for Pyxel, recently used it for a couple of hours to create a tiny project [1] and loved how intuitive it was.

[1] https://github.com/abhishekbasu/minesweeper

This was fun. I saw a post on Pyxel a couple of days ago, and decided to write mine in Python using it.

https://github.com/abhishekbasu/minesweeper

There are [1, 2], and AFAICT they rely on gnuplot [3].

[1] https://github.com/nschloe/termplotlib [2] https://github.com/dkogan/gnuplotlib [3] http://www.gnuplot.info/

your pessimism is hardly unwarranted. this will lead to rather lousy code documentation, and might be misused by folks. but on the other hand, i look at this as something that will be effective as a "summarizer" of sorts where it describes already written code with lack of/poor documentation.

I wonder if this can somehow be used as a tool for quick visual inspection of architecture compliance. Maybe the size of the box could be selected before display to denote a custom metric. For example, if the size of the box for each file was proportional to the (number_of_lines_edited) * (current_date - file_creation_date), then modularity could demand that the size of the boxes remain small. (pardon the musings of a non software engineer on a Saturday night)

While HTML and other markup languages are great, I wonder if there is a way to use a word processor while creating the document - focusing only on content (spellcheck, etc.) and not on formatting or on syntax. Then, a context specific converter (that is specific to the type of content - a resume, a newsletter, a FAQ page, etc.) takes care of adding the formatting and publishing it.

As an example, recently, I had to create a bunch of FAQ pages, and I created a tool like [1]. While Pandoc can be used, I still need to worry about (at least basic) formatting the doc while typing it out in the word processor to make it look good in the HTML output.

[1] https://github.com/abhishekbasu/faq2html

7380 with one forced error (was cooking while answering this). Can't agree with a few in the medium difficulty though, felt that the "correct" answer was arbitrary. However, IANAD so perhaps not the best person to say for sure.

First off, congrats on your venture! as a foodie I'm elated.

I realize that you're trying to create your own brand, my concern is:

- if it's a brand that does all cuisines, it is going to attract people mainly because of the price point, and not uniqueness

- if the idea is to build multiple brands, one each for a specific type of food: a brand for Pizza, another for Biryani, etc., then scaling each is its own demon

please correct me if I'm not understanding it right.

On a side note,

> But it didnt work and they shut down all locations

Do you have any knowledge of why it didn't work? I have a few thoughts around this, and have discussed this with a friend who's a restaurateur, but would love to hear from you!

On a side note, if the facial expressions were spot on, how would you as a prospective customer feel if you were later informed that the video was in fact AI-generated?

Pardon me if I'm not thinking about this in the right way, but how would you hold the store owner accountable for refreshing or updating their catalogue on the app if they don't already use an inventory management system for purchasing and re-stocking? Isn't that a huge shift in behavior?

I often wonder if LI is trying to do too many things at once. At this point it hard not to come to the conclusion that it promotes certain types of behaviors. The problem is that if left laissez-faire, the feed will not attract folks who abhor work influencers. So, perhaps they need to ask who they are really targeting?

For generic work-banter, an app like Blind seems to cut it better. Given the anonymity, it is able to (anecdotally) elicit truer depictions of one's workplace, and this helps folks get real value through what are in essence (unaggregated) reviews of companies and job roles.

By using real profiles and establishing a set up of "professionalism", LI absolutely cuts/reduces the possibility of deriving real value for the layman (read non work influencer and non CXO). Apart from connecting with old colleagues and looking for a new job, why should the layman even bother logging into the platform on a fairly regular basis?

Having said that, LI is great for some things even for the layman. It has been great for looking for new jobs as the article states. But it seems to have a crisis of identity, and it needs to figure out their core audience or differentiate product offerings before it deteriorates into a job board for the laymen like me, in which case it will be possible for a new and trendy job board to come along and replace it.

I'm not a UI/UX guy, but the presentation in the book was so good, I decided to read it all in one sitting! As a user, I have faced the problems you describe in pages 38 and 57, websites should always try and make it clear where there are more options/content that may be hidden due to lack of screen real estate.

With AI-based support automation, it might end up like https://www.youtube.com/watch?v=WnzlbyTZsQY