Shreyash

Software Engineer | Full-Stack | AI and Machine Learning

Building something that excites me 🔥

[Shreyash @WebDevCaptain]

@sebastienlorber @rickyfm 👀👀👀

[Shreyash @WebDevCaptain]

@dharmeshba Security >>> Animation

[mem0 @mem0ai]

https://t.co/cQ9EO2FSzc

[Shreyash @WebDevCaptain]

@sunnykgupta @deepdiwakar9 Aur khatam hote hi wapas aa jate hain 😂

[Shreyash @WebDevCaptain]

@Anand240239 @ArvindKejriwal थे

[Shreyash @WebDevCaptain]

Will anything really change just by constantly stepping back ?

You are educated (although somewhat authoritarian at times), and if you truly have the courage, then fight across the entire country. Go to Varanasi and challenge that "4th-pass" leader. But no - you prefer to play it safe, slowly build a narrative, and when the time comes for a real fight, you focus only on places where victory is virtually guaranteed.

Sir, you are no longer the person who once gave hope to the youth.

If you don't have the courage to do it yourself, then project Manish Sisodia, Atishi, Saurabh Bharadwaj, or Sanjay Singh as the Prime Ministerial candidate for 2029. Don't just make speeches - take action.

#IndiaNeedsEducatedPM

[Shreyash @WebDevCaptain]

@rezoundous Nah, they will become obsolete very soon...

[Shreyash @WebDevCaptain]

#MinimaxM3 is definitely an upgrade over m2.7 but it's definitely not comparable to GPT 5.5 or Opus 4.6

I still believe that Minimax and GLM have the potential to become the new frontier models

[Shreyash @WebDevCaptain]

@mehulmpt India is losing good talent but our policy makers are busy in dividing the country and wasting youth's time & energy; Sad 😞

All the very best Mehul

[zack @zack_overflow]

Why does tsgo use so much memory?

Because it literally never frees memory allocated for types!

And multi-threading overhead makes it likely to duplicate typechecking state on multiple threads

I wrote about this in more detail here: https://t.co/7od2GMrVAW

Basically:
- nearly every type is allocated and then never freed (including transient types which are created and destroyed very quickly)
- each thread gets its own type-checking state (types, symbols, etc.)
- but they don't share data
- so it's very easy for one thread to end up recomputing and allocating memory for types/symbols/etc. that another thread already did

The end result is a very high peak memory usage!

[Shreyash @WebDevCaptain]

@drummatick Linkedin premium is just useless if I exclude Linkedin Learning and unlimited Note feature on connection request

Search is broken always

[Shreyash @WebDevCaptain]

It's slowly building up, will crush Anthropic & GPT in next 1 year for sure

Don't forget Google DeepMind

[Shreyash @WebDevCaptain]

@arpit_bhayani Great explanation, very rare to find these kind of indepth videos on YT.

1 req, Release 2 more today if possible.

[Shreyash @WebDevCaptain]

@Venkydotdev @sama

[Shreyash @WebDevCaptain]

@arpit_bhayani ++

[Shreyash @WebDevCaptain]

@AlexanderAqvt7 @haider1 IDTS, I am using it continuously and don't see any improvement compared to 4.7; tool calls are failing more often

[Shreyash @WebDevCaptain]

@michalmalewicz Best strategy to pump your stocks 😂

[Shreyash @WebDevCaptain]

@justjayvi @theZachChen