This looks like a very exciting initiative.
It covers 2 major pain points for gold loan industry:

1. The opex in evaluation is a considerable expense for the lender
2. The borrower has to spend considerable time physically at the branch for disbursement.

https://www.business-standard.com/content/press-releases-ani/goldpe-launches-apm-automated-pawn-machine-a-revolutionary-ai-based-gold-loan-atm-123103000811_1.html

“With GoldPe | APM, they can get their gold loan instantly, safe and secured within less than 8min & 24x 7. Customers can deposit their gold in the machine, which then uses AI technology and Gold / Metal Analyzer to assess the Purity, Quality and Weight of the gold. The AI system will weigh the stone and gold weight separately, and based on this assessment; the machine instantly disburses the loan amount”

AP train crash: Full Kavach cover may take a decade | Mint (livemint.com)

• The official said the railways plans to float a ₹5,000 crore tender early next year for fast-tracking the adoption of Kavach. These include two tenders for Kavach, on a 6,000-km network each. This is double the 3,000 km first tender for Kavach issued in 2021.
• According to an earlier plan of the railways, top priority would be accorded to high-speed trains on busy routes in the New Delhi-Mumbai and New Delhi-Howrah sections.
• The second priority was to install Kavach on trains running on heavily used networks with automatic block signalling and centralized traffic control.
• In the third phase, Kavach would be introduced on other high-density routes with automatic block signalling.
• Officials said the priority order will now be fast-tracked for this anti-collision system, which works best for heavy-density routes where trains run at frequent intervals and the distance between two trains is often a lot less than on other routes.
• Other than Kavach, automatic block signalling will stop trains automatically if the driver fails to see a signal. Centralized control would add another level of manual or physical check on train operations supported by system checks.
• So far, around 1,445 km of network has been covered with Kavach. With new tenders, the initiative would be expanded quickly along entire railway network of around 70,000 km.

India’s ‘Kavach’ Protection For Trains: Could This Be The Formula To Run High-Speed Trains? – News18

Ericsson showcased the power of LTE and how it positively impacts the performance of the Kavach system and reduces the latency to ensure that trains can run at high speeds in a secure fashion. The interesting part about Kavach is that it can be integrated with both new and old-gen trains in the country as it is backward compatible, which saves Indian Railways a lot of money which would have been needed to upgrade the existing trains to equip the Kavach security system.

a short youtube video link embedded – Indian Railways Kavach System Explained: How This Anti-Collision Tech For Train Works – YouTube

a few points from the above video link:
right now working in UHF Ultra high frequency and if it moves to LTE.

• it takes 4 seconds to register a locomotive in UHF v/s 160 milliseconds in LTE
• it takes 2 seconds to communicate between Kavach and Station v/s 40-50 milliseconds in LTE

it helps to run trains at higher speeds when on LTE, especially like Vande Bharat trains.

They should be able to do 1300+ cr PAT in FY24 if the status quo is maintained w.r.t. credit costs. H2 sees higher growth relative to H1. I expected a bigger reduction in the off-roll collections team this year, but that lever is yet to play in decreasing the OPEX. The management guided this, but the quantum of the reduction was never mentioned IIRC. Cost to income has already reduced even without this lever.

Honestly, I expected the PAT to normalize, and stagnate in FY24 itself, but the credit cost cycle had a mind of its own. I doubt anyone can predict when the normal will return. All we know is it will.

So, while it can happen, but H2 is not when the credit costs looks to normalize. If you go through the con calls of all MFIs, SFBs, they are all guiding 1-1.5% credit cost this year. Some are foolish to guide for next year as well, they can be ignored.

So how does one play the Vedanta story from here?

It’s a given that the promoter Anil Agarwal does not hold minority shareholder interest anywhere close to his heart. The hefty dividends paid were due to the promoter’s need for cash to service the debt. It is also a given that no meaningful investor will invest in the Company in its current avatar. The Company did sell some stake in the recent past n the share price promptly took a drubbing! So any further stake selling by the promoter is not really an option for want of any takers, But therein lies the opportunity. Vedanta, as already proposed by the Co. recently, (link attached below) will have to be split up into many focussed commodity specific smaller companies & sell controlling stakes in a couple of them to raise any meaningful capital that could take care of the stretched balance sheet. The Co. is sitting on a number of prized assets which would find numerous takers if the current promoters are out of the equation! That seems to be the only possible way out for the promoters to get out of this debt trap.

If that were to happen there could be substantial gains for investors to be made over the next 12 months or so. The current price levels of about 215 appears reasonable to enter, but the basic investment thesis assumes that the demergers will go through.

(For some reason there seems to be two threads for Vedanta)