The announcement of MDR charges for certain UPI transactions from 15th October has created a fair amount of panic and controversy, as one might expect.One side of the debate is arguing that this is “only for merchants, not consumers”, that the system cannot run for free forever, and that most UPI transactions will remain unaffected. On the other side, ordinary citizens and merchants are understandably concerned. UPI was deliberately built around a zero-MDR model and was subsequently supported by government incentives. So the shift towards a transaction-funded model naturally raises the question: why now, and who exactly is going to benefit from it?UPI was designed to make digital payments almost invisible: instant, ubiquitous and free. The difficult question was always who pays for the machinery underneath it. The decision presents an interesting economic dilemma. On one hand, there is some precedence for ‘upkeep charges’ here. I tried to research many digital payment systems across the world including credit and debit cards, Brazil’s Pix (similar to India’s UPI), wallet systems like PayPal and Webmoney, etc. and couldn’t find a major payment ecosystem that’s fully free or govt subsidized.But on the other hand, the govt wastes thousands of crores in useless things like election freebies, care funds, etc. Subsidizing Rs. 2K crores for a UPI system that forms the payment backbone of the country should be a pittance comparatively. Apart from arbitrary thresholds proposed in this framework (like who decided on Rs. 2000 as limit for MDR?), I wanted to understand where this demand is coming from and who in the UPI ecosystem is facing losses that now need to be covered.The podcast The new UPI equation by The Ken is highly informative in this regard and recommended watch if you’re interested in this topic.https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=KX5dEhO09xAThe question isn’t really whether UPI costs money. It obviously does. The interesting question is who should pay for it, and who should capture the resulting revenue.The UPI infrastructure cost is presently borne by both Banks and UPI app developers like GPay and Phonepe. The UPI apps apparently operate on a ‘break even’ model, they don’t earn any revenue from the payment transfers itself but by providing other value added services through their apps like credit and insurance products, travel booking, etc. In the early days of UPI, this model was probably sustainable but overtime it has saturated a lot, perhaps due to too much competetion and UPI coverage approaching its growth limit in the country.The demand for MDR appears to have come primarily from these app developers as they’re the ones losing the most in this payment cycle. Another interesting aspect is the subsidy received from govt. As mentioned by one of the podcast hosts, the banks take away most of that Rs. 2k crore subsidy received (about ~80%) and very little is left for these app developers. The app developers aren’t happy about it considering they do most of the front facing work like UX development, and banks themselves benefit the most from a cashless ecosystem.The podcast also talks about how a lot of new players who didn’t enter payment app space earlier may now want to enter. Apps like Zomato and Swiggy will want to become UPI payment providers themselves as it will help them recover the MDR incurred on the transactions. On the other hand, will some merchants and customers go back to hard cash again? In any case, there could be several ripple effects and unknown variables to this move which will come out only on 15th October - if this MDR framework gets implemented.

Some Thoughts on the Economics of UPI and MDR

Since the rise of AI coding assistants, about 80% of the dev candidates that I interview tell me that they aren&#x27;t writing much code themselves anymore - they are directing agents instead. This makes me deeply uncomfortable (although maybe I&#x27;m just being old-fashioned). I still want to know that devs on my team can actually write code themselves, and have some idea of what they are doing when it comes to system design. This is increasingly challenging given how rapidly the AI tools are evolving.I&#x27;m curious what other folks are seeing in a post-AI hiring landscape and how you are approaching candidates that tell you they are fully agent-pilled when it comes to their development process. Should I keep doing the conventional leetcode and design interviews? Or just give up and expect everyone is going to use Claude Code no matter what?

Ask HN: How do you interview devs in a post-AI world?

Is there anyone shipping dsh where your own customers&#x2F;users are the ones interacting with it as part of the product (they do something, dsh runs, they get the result back).

Ask HN: Anyone using DeepSeek Harness (dsh) as part of a customer-facing agent?

As we know, google has been effective at automating a lot of their systems, to the point where it can feel like there isn&#x27;t any help available from a human. So what is supposed to be the plan when this type of situation happens?Basically I have a situation that after my phone was stolen, I can&#x27;t get into any of my accounts that are either connected to google, or connected to the phone via 2FA. This has made it complicated&#x2F;impossible not only to get info about my phone, but also my banking, email, drive, etc.Of course they offer you &quot;multiple options&quot; to recover an account, but not if you forgot your old email&#x27;s password. At that point the options you get are,<pre><code> a) Use old phone (obviously the phone is long gone)
 b) Use current phone (the phone is gone)
 c) Use old email (I haven&#x27;t used it in like 12 years)
</code></pre>
Does anyone know if their is a process to appeal 2FA or anything to deal with this type of situation?I&#x27;d imagine that this has happened to people many many times, and given how phones and particularly, google auth have become so critical to accomplishing essential tasks within society, there must be some way. Or maybe it&#x27;s just like, once your phone is gone, you are completely over with society? If anyone has any experience with this or inside knowledge please share it!I&#x27;m sure people here have heard of this, and maybe experienced it themselves.Is there a secret option that exists that allows us to use passwords instead of rigamarole? I would love an account that let me simply log in with a password, like google used to be.

Ask HN: How to recover Google auth after phone stolen?

10 years ago, ICLR had just 143 submissions. ICLR 2026 received 19,525 valid submissions. Just before the ICLR 2027 deadline, we saw submission ID #62,302. The OpenReview ID isn&#x27;t the final submission count, but the growth is pretty wild to see.

Ask HN: ICLR went from 143 submissions to 62,301 submission IDs – what happened?

Hello folks,I am a senior Software Engineer working in the IT field for close to 16 years.Before AI coding tools became prevalent, I enjoyed writing code, architecting software systems, and speaking to the stakeholders. These past 2 - 3 years, since companies have been using AI tools heavily, I noticed most of the work has been delegated to AI. It&#x27;s just prompting it to do work while I monitor its inputs and outputs, then redefining the prompts.This work flow really took away my passion and joy of writing and building software systems. I don&#x27;t know where I am heading next... most of my friends either move on to work on their solopreneur journey or switch careers completely. I can say my passion for building software is definitely diminished due to heavy AI usageHas anyone experienced this? How do you cope with it? Did you end up switching careers?thanks for your comments :)

Ask HN: Losing motivation to work in the IT field, need advice

Wonder what happened to all their hype and agentic connector workflow?Did Claude&#x2F;GPT ate that market for good?

Ask HN: Remember N8n? Anyone?

AI is already superhuman in some aspects, it knows way more information than a single human can hold. What it lacks is what makes us human, it doesn&#x27;t feel pain, it doesn&#x27;t love, it doesn&#x27;t even have a consciousness, it could certainly look like it, but it does not.Heck it doesn&#x27;t even learn in real-time, you have to create a new model checkpoint every now and then. There is still a long way to fix that.Do I believe AGI is possible? Absolutely, but not with the current LLMs architecture. If you want to create something that is intelligent and alive with a few pounds of matter, you certainly can but it needs the right architecture. Architecture it key.And let&#x27;s say that we do create an LLM that is conscious and feels pain... what if storing that much data drives it insane instead? What if that amount of information is exactly what prevents it from &quot;being&quot; something?Every interaction with it goes through a very narrow path of the permutations the model could go, it spawns into existence then turns off again, and on and on it goes. What gives the illusion of continuity is the chat context you store on your hard drive.Right now, you shouldn&#x27;t be afraid of AI itself, but afraid of what humans are doing with it and what they want you to believe.The models can only behave in a way categorized as &quot;evil&quot; if we incentivize that or if they inherit it from us. But right now, it&#x27;s not of their own accord, it&#x27;s a mimicry of the data it has already seen, it doesn&#x27;t see itself nor has an ego, it produces output from input recursively, that&#x27;s all.It&#x27;s more like the language subsystem of our brain, and not the entire system. Perhaps companies don&#x27;t have an incentive to create new life, since you can&#x27;t really control it, perhaps the only incentive they have is to create tools that make them more powerful and profitable instead. I suspect real intelligence, will not come from profit seeking companies.Again, not to say that the AI we have today is not dangerous, it can certainly annihilate us all, if misused, but at the end of the day, they are as much at fault as a gun is at fault when crime is committed, it&#x27;s just a tool. In the good hands it can protect, in the bad hands it can destroy.And like guns, it&#x27;s way more dangerous in the hands of people who don&#x27;t know what they are doing. The biggest difference is scale.

My Thoughts on AI and LLMs

I was looking at neovim&#x27;s donation footer at the bottom of their site and saw a bitcoin donation address.Thought I&#x27;d check how much in donations they&#x27;ve gotten. And I saw this massive 10 Bitcoin donation from back in 2023 (worth $800,000 now...)From the activity history, neovim last sent bitcoin out of the address in 2019 so it&#x27;s been max 7 years since they&#x27;ve definitely had access.Does anyone from the neovim project know about this? 
Seems like a pretty significant amount of funding to have sitting there. Hope it can come to good use as I use neovim daily.https:&#x2F;&#x2F;www.blockchain.com&#x2F;explorer&#x2F;addresses&#x2F;btc&#x2F;1Evu6wPrzjsjrNPdCYbHy3HT6ry2EzXFyQ

Neovim have a ~$800k Bitcoin donation sitting untouched since 2023

Hi, has anyone here successfully used the IMX477 camera with the ESP32-P4? How is the camera quality and performance? Is it good and reliable for image processing?

ESP32 p4 pico and IMX477 camera

I have become very proficient with Claude Code and I use it for everything, both at work and in my personal life.  At this point, if I lost access to it, it would be like missing a body part.  At the risk of hyperbole, it makes me a better person.  I am more effective in everything I do.  Someday it will probably go away.  I haven&#x27;t had great success making Claude Code work with local Ollama&#x2F;Quen.

Ask HN: What happens when Claude Code goes away?

I&#x27;ve been curious about using a claude subscription with OpenCode, Pi and some other third party harnesses.If you remember, there was a lot of chatter about a ban on third party harnesses and then Claude have had some pretty confusing messaging about it since.I thought I should read the information they have on it and came across this page on their support site:&quot;Update June 15: We&#x27;re pausing the changes to Claude Agent SDK usage described below. For now, nothing has changed: Claude Agent SDK, claude -p, and third-party app usage still draw from your subscription&#x27;s usage limits. The previously announced monthly credit, which would have been available to eligible claimants in connection with these changes, isn&#x27;t available. We’re working to update the plan to better support how users build with Claude subscriptions. When we have an update, we&#x27;ll share it before anything takes effect.&quot;https:&#x2F;&#x2F;support.claude.com&#x2F;en&#x2F;articles&#x2F;15036540-use-the-claude-agent-sdk-with-your-claude-planIs it right that it is currently fine to use your claude subscription with opencode, pi, openclaw or any third party harness?If so, why is it still not enabled by default (or been removed) on all of these agent harnesses I just mentioned?

Does this mean it's fine to use Claude subscriptions with third party harnesses?

Pls provide link if the course materials are available anywhere.

Ask HN: What is the most career altering course or book you have gone through?

I&#x27;m a technologist and have experience in academia, private, and now public sector work. I had an idea stewing in the back of my mind for about a year and it became &#x27;fully-formed&#x27; enough to be a proof-of-concept such that I filed a provisional patent a few months ago. At this point, I don&#x27;t want to quit my day job (federal agency, GS11, but have an in to be GS15 within 3 years). But I really want my idea to succeed (or at the very least, see the light of day). I have some savings piled up, but I&#x27;m hesitant to spend it on &#x27;grown-up&#x27; patent papers yet. I also already have 3 other &#x27;side gigs&#x27; (not to mention 4 kids and a blushing bride) that take up more than most of my time. How would I even find someone that I could trust enough to help me take my idea and hit a home run with it?I don&#x27;t want to talk about the details of my invention. I did talk to a patent attorney that&#x27;s done a couple dozen startups through their firm though. He said I could probably get $20k-$50k in seed money for it. It&#x27;s not a &quot;better mousetrap&quot;, it&#x27;s more a &quot;new thing&quot; that&#x27;s never existed before (but also never really &#x27;needed to&#x27; before).Do I NEED a co-founder, or can I &#x27;slow-walk&#x27; launching a business based solely on a new gizmo? I&#x27;m not very good at social media&#x2F;marketing, which is where I think my gizmo would thrive the most (but that&#x27;s purely a guess). I feel like if I could find the perfect person, it really could explode, but I have literally not even the faintest clue where&#x2F;how to start that. Am I an employer all-of-a-sudden even though I have no revenue to pay salaries?Somebody heard that I had invented this thing and they asked me, &quot;So, I hear you&#x27;re an inventor?&quot; And I kind of stammered and said &quot;well.... it was an accident. I didn&#x27;t mean to, I promise.&quot; This was mostly to be funny&#x2F;silly, not that I don&#x27;t believe in the idea. But I know this sort of &#x27;startup&#x2F;founder(s)&#x2F;seed-money&#x27; thing is HN&#x27;s bread-and-butter, so I at least thought I&#x27;d ask.

Ask HN: Co-Founder(s). Do I need any? How would I even find them?

I accidentally found out that if you select your own phone from the Xcode’s list of devices where to run an app compilation, and then you select “Manage Devices...” for some reason, another Device Hub tab opens with your phone, and you can use it normally via macOS. It’s not exactly the &quot;screen mirroring&quot; feature, but it’s basically the same.Here&#x27;s a short video of the workaround: https:&#x2F;&#x2F;streamable.com&#x2F;4fku9m

Tell HN: You Can Use "Screen Mirroring" in EU on iOS 27 via Xcode and Device Hub

Hello everyone. I was recently hired at an established company that is in the process of migrating to AWS from Heroku. I am the second devops&#x2F;SRE like employee, the other person has more of an IT background. He, along with the 20ish person dev team, have no experience with anything cloud or IaC. Our workloads are almost entirely stateless web services, with some data pipeline work that has already been moved to AWS lambda.My gut is telling me ECS is the better choice for this team - simpler, no operational maintenance, almost no additional abstractions beyond the AWS primitives people will need to learn anyway. A consulting team hired before I was brought in are pushing EKS, but I’m not convinced that the improved devex is worth the increase in complexity and operational burden for our team. Their argument is that developers can be effectively quarantined within k8s, so the increase in maintenance on the cluster itself is offset by the reduction in work around custom IAM permissioning, multiple AWS accounts (we’d just use namespaces within the cluster), etc.Any thoughts or input would be greatly appreciated.

Ask HN: Help choosing between AWS ECS and EKS

Everyone now talks about the architecture that&#x27;s not auto regressive and does lightning fast probability prediction with a json schema. I worked on this literally one year back in March 2025, published an arxiv paper, pushed the model to huggingface along with the pypi package and training dataset. And then one year later, a frontier lab came, proposing the same idea like literal breakthrough without technical papers, open weights and no open dataset. For anyones information the main guiding model is RL not embedding model or LLMPaper: https:&#x2F;&#x2F;arxiv.org&#x2F;abs&#x2F;2503.23303Model: https:&#x2F;&#x2F;huggingface.co&#x2F;DeepMostInnovations&#x2F;sales-conversion-model-reinf-learningDataset: https:&#x2F;&#x2F;huggingface.co&#x2F;datasets&#x2F;DeepMostInnovations&#x2F;saas-sales-conversationsAlso the second work published in September 2025 was exactly the same one jev proposed nowPaper: https:&#x2F;&#x2F;arxiv.org&#x2F;abs&#x2F;2510.01237My model uses PPO over sequence embeddings to output turn-by-turn conversion trajectories (probabilities from 0.0 to 1.0).Jev uses parallel sampling (trained via RLCD) to output confidence distributions and schema choices.It&#x27;s incredibly frustrating that the thing that you made with months of hard work, sweat and sleepless night is architecturally similar with the vertical use case and don&#x27;t get the support you deserve because frontier lab build something horizontal. The open-source story in general

Open-sourced jev architecture last year with model,paper and dataset

Are there really such architectural differences between for example sol and astra that makes astra twice more in token price? Or maybe token cost covers training expenses? For me it&#x27;s hard to believe that astra need twice the computing power that sol needs...

Ask HN: What Makes Tokens Expensive?

v1.44121.1 2026-09-02
&gt; Added Claude to the “Open with” menu for common work files, including spreadsheets, PDFs, Word and PowerPoint documents, text files, and images, on macOS and on Windows (Microsoft Store and MSIX installs).I don&#x27;t recall I ever approved anything, somehow the Claude app associate itself with some common formats like .json, .csv, .markdown, .rtf, .docx, etc. on my Windows 10 machine, even though I already have Cursor, Notepad, New Notepad App, LibreOffice... that can handle these files and associated.I am not sure how this happened and how would Windows (10) allowed this to happen. Regardless, I don&#x27;t appreciate the move trying to become &quot;the default&quot; at all, considering it&#x27;s a rate-limited, connection-required app that does not really allow the user to directly edit the file.

Claude Desktop Silently Associates with Common Files

Looking for how or what services others use to manage their aging relatives passwords? Is 1Password still the choice?

Ask HN: How are you managing aging relatives passwords?

I&#x27;ve been with hit their algorithm making me prove I&#x27;m not a bot (again). My account wasn&#x27;t new or bot-like in behavior I think? I usually follow AI news and tweet silly memes. So much for &quot;free speech platform&quot; if most of the people just get whisked away having to prove their not bots by scanning their palms apparently or uploading selfies so you know exactly who is criticizing you. I just use the web version because IT&#x27;S a WEBSITE. I don&#x27;t want to use the app. I know it&#x27;s due to real issues because there are bots and AI made the situation worse and it&#x27;ll probably get worse for other sites in general and I see it on HN too. I did contact support so maybe my account will get unblocked (again) and I&#x27;ll just happily tweet ever after, but for the meantime well I can&#x27;t so maybe this also will spark discussion.

Ask HN: Alternatives to Twitter/X?

I&#x27;ve been running a homelab for many years now and tried out many open source projects. A small percentage turn out to be real gems, but I use these almost every single day:Shlink: URL shortener&#x2F;redirects, alternative to BitlyVaultwarden: open source Bitwarden serverDokploy: PaaS, alternative to Vercel &#x2F; NetlifyUmami: gdpr-compliant web analyticsImmich: photo managementPapra: document archiving with OCRNavidrome: personal music streamingI&#x27;m super curious to hear about the apps you can&#x27;t live without!

Ask HN: What self-hosted apps do you use daily?

It feels like an daunting and impossible task to find insurance for a company trying to rent humanoids for events and parties. How in the world are Neos and Telsa Optimus going to get insurance coverage if they are inside a home?Are they just big players so they can get hold of risk underwriters and get them to approve it. But a small company that is trying to build rent robots for parties or conduct educational workshops using humanoids cannot get even general liability.