WHY SWALPA? (https://swalpa.org/behind_swalpa/why/)

This section dives into the "Survival Kannada" philosophy. It explains why we prioritize high-frequency scenarios like negotiating auto fares or understanding local etymology (e.g., why tech hubs like Marathahalli still carry the "-halli" village suffix) over standard textbook grammar.

AI ARCHITECTURE & WORKFLOWS (https://swalpa.org/behind_swalpa/architecture/)

This page breaks down the "Local-First" stack and the agentic loop:

* The Staging "Brain": I maintain a private repository where I use Gemini Deep Research to crunch raw linguistic data. This verified "Spec" is what I then feed to Antigravity (the coding agent) to build out the game engines and lessons.

* Portability: The framework is designed to be "forkable." A sister project for Tamil (konjam.org) is already being built using these same architectural principles.

I am happy to dive into the weeds on the prompt engineering, the TTS pipeline, or the Bangalore-specific nuances if anyone is interested!

I built SWALPA (https://swalpa.org) because learning spoken Kannada is a New Year's resolution for me and I realized most language apps teach you formal grammar, but not how to negotiate with an auto-rickshaw driver or handle the "daily chaos" of the city.

THE AGENTIC STACK (BUILT WITH ANTIGRAVITY)

I wanted to see how far I could push a "Human-in-the-loop" AI workflow to create high-quality content in a low-resource language.

* The Coder: Antigravity (an agentic coding AI) handled the heavy lifting—writing the JS game engines and drafting the core lessons based on a custom "grammar spec."

* The Knowledge Base: I used Gemini Deep Research to build a high-fidelity reference of Kannada linguistics. This acted as the "Source of Truth" to prevent AI hallucinations.

* The Content Engine: I fed my core curated lessons into NotebookLM to generate supplemental podcasts, slides, and flashcards to boost recall.

* Architecture: Simple and Boring—Static site (MkDocs Material) + Vanilla JS + Firebase.

THE EXPERIENCE

Instead of "The cat is on the table," SWALPA uses interactive games to simulate real-world Bangalore scenarios:

* The Auto-Rickshaw Negotiator: Use the right phrases under timer pressure to get the driver to use the meter ("Meter Haaki!").

* Rapid Translation: A high-pressure mode where you hear a Kannada phrase and must pick the correct translation from four options under a timer.

* Agglutination Mastery: Interactive drills for Kannada's unique word-stacking grammar.

The platform includes 10 core text lessons and Duolingo-style gamification—including badges, streaks, and an activity heatmap.

Games Link: https://swalpa.org/games

Source: https://github.com/saurabh-net/swalpa

I am a Software Engineer at Google, but this is a personal passion project. I would love to hear your thoughts on the AI workflow or the "street-smart" approach to language learning!

Comments URL: https://news.ycombinator.com/item?id=47215918

Points: 4

# Comments: 2

But their stock price keeps going up and they keep eating up new markets.

But on the other hand, China was forced to agree to the HK situation particularly since they were at a disadvantage militarily and economically to the British empire.

China (CCP) would understandably not be in a mood to comply with agreements they felt coerced into.

I genuinely don't understand the full situation, and I'm not trying to imply that the CCP is in the right.

I'm just trying to understand if the whole "bullied into a disadvantaged agreement" is a valid world view to take.

Everything the CEO mentioned last time is still applicable.

What's worse is I'm not sure we'll be able to collectively drum up enough outrage to get it reversed this time.

Got me a little nostalgic, and it's such a joy to read pages just focussed on the content and absolutely nothing else.

The search results are sometimes bewildering, I guess search quality could be a little better. But on the whole, a lot of fun.

Thanks for sharing :-)

"Huawei was able to eavesdrop on Dutch mobile network" in the original sounds a lot less incriminating than "Huawei eavesdropped"

I wonder if a part of that longevity is the org/incentives structure.

And (cynically) maybe just that regulation entrenches existing finance players and there isn't anywhere near as much anti-trust conversation over there.

FWIW, I absolutely love the kinesis. I have 2 pairs and the older one is 4+ years old.

It helped me hugely with RSI around my wrist and fingers.

The other thing which really helped was classic strength training at a gym :-)

The general consensus seems to be that powerful type systems are very effective.

Personally, the low footprint runtime and concurrency primitives are enough for me and I wouldn't mind the language becoming "less simple" if it helps the ecosystem.

Once generics are implemented, I can imagine people requesting for the next "missing" thing.

The original title definitely sounds more inflammatory than the one they changed it to later.

Thanks for clarifying!

And it's deliberately titled to make it sound worse than it is (at the moment at least).

I'm not sure how one requests a title change on HN, but that would help here.

I'm not surprised multiple vendors haven't adopted the system yet.

2.2 million on a research initiative and not immediately adopting it until it's proven and widespread seems perfectly reasonable to me.

The (abridged) title made it sound worse IMO.

The general healthy advice all worked for me to different degrees.

Exercise, meeting friends and having non-desk/screentime intensive hobbies.

Meeting friends and nurturing those relationships helped hugely. And sadly, the second most effective thing was switching to a less-intense team.

I've definitely seen other friends with unhealthy habits and anxiety in the industry too so I'd try some of these and see if it helps.

And you could definitely try therapy and/or medical help too.

https://youtube.com/playlist?list=PLrw6a1wE39_tb2fErI4-WkMbs...

It's easy not to care since Parler is the "bad guy" here, but I do think that Internet infrastructure companies need to give a reasonable heads-up before pulling the rug under business customers.

They do paper reviews and I pick the papers/topics which seem the most interesting (rather than doing them in order).

They are great if you've already gone over the basics using "Designing Data Intensive Applications" or "Distributed Systems for Fun and Profit".

https://www.youtube.com/playlist?list=PLrw6a1wE39_tb2fErI4-W...

Especially with remote work on the rise.

This is one thing I hope all of us continue opposing vehemently.

It's a huge privacy concern and companies have too much incentive to want to monitor this kind of data.