Dipendra Bhatt

SSE 2 at @harnessio • Previously SDE 3 at @hiverhq and Founding Engineer and Backend Team Lead at @mydukaanapp
Databases, especially postgres excites me.

[Dipendra Bhatt @dipenbhatt03]

@mholt6 @caddyserver i am genuinely surprised people havent heard of caddy, it has always been my goto when i didn't wanna deal with nitty gritty of setting up nginx 🤷🏼

[Anish De @AnishDe10]

Today, the wedding of a known person was called off because the boy’s family came to know about the girl’s multiple affairs. She also demanded to live separately from both sets of parents, and asked him to take an internal transfer abroad so that they could become NRIs.

Ab tak sirf social media pe sunta tha main, ab real life mein bhi dekh liya. Scary times, fr ☠️

[Planet PostgreSQL @planetpostgres]

Haki Benita (@be_haki): Unconventional PostgreSQL Optimizations https://t.co/SZUaqzeiY5

[Wise @trikcode]

British Python devs be like "that's a constructor, __init__ ?"

[Dipendra Bhatt @dipenbhatt03]

At the same time there will be those particular cases, where rawdogging SQL is the way to go to squeeze out maximum perfomance. As always with everything in engineering, use the tool for which it is best rather that considering one size fits all.

[Dipendra Bhatt @dipenbhatt03]

Cannot agree more with this, discarding a good ORMs with the only argument being we have hit scale is preposterous. Any complex system, will have a mix of simple queries (where writing raw queries is just over kill).

[Valery Melou @valerymelou]

I've worked on systems serving millions of requests, and this binary choice fallacy doesn't hold up in reality.

At scale, 95% of your queries remain boring CRUD (Get User by ID). Writing raw SQL for these is wasted engineering hours. You don't migrate away from the ORM. You keep it for the heavy lifting and drop to raw SQL only for the 5% of complex bottlenecks

Also explicit queries sound nice until you need to rename a column or migrate a schema. With an ORM, it’s a code change. With raw SQL, you are grepping through thousands of string literals praying you didn't miss one. That is tech debt.

Dismissing the importance of ORMs, even at scale is a mistake according to me.

[Dipendra Bhatt @dipenbhatt03]

"In the midst of winter, I found there was within me an invincible summer".
This line fits perfectly for you AP @NeurlAP . Proud of you!

[Abhinav Prakash @NeurlAP]

3/n
Started applying for remote jobs even during my father's last rites.
Got selected at Pocket FM as a Software Engineer in the Gen AI team.
Built multiple things from scratch in Gen AI, and today got promoted to Senior Software Engineer.
What a wild ride. 🚀

[Dipendra Bhatt @dipenbhatt03]

Damn, just yesterday i was dicussing with my peers, where one of the potential solution to a problem was self hosting LLM due to privacy concern.

[Nic Conley @niconley]

this guy is getting paid $35,000 to set up an internal “ChatGPT” for a law firm.

> locally hosted Llama for LLM
> N8N to connect it all.

we’re living in the AI gold rush.

[kitze @thekitze]

facts

[Dipendra Bhatt @dipenbhatt03]

Or always use a read only user for Prod db for regular tasks.

[Jason Bosco @jasonbosco]

NEVER open your production DB and staging DB in two different tabs simultaneously (CLI, GUI, or whatever you use to access it).

Always close one, before opening the other.

IFYKYK

[Arnav Gupta @championswimmer]

So I have already started discovering a pattern about how "amazed" people are with LLMs generating code.

1. PMs, non-tech founders, wordcels and yappers are the most amazed. They can sit on their couch and write "make me an app" on their iPhone, and it makes them yet another landing page. They cannot stop waxing eloquent about how AI is changing the world.

2. Indie developers, script kiddies, and more junior developers also love "vibe coding". They keep saying "Fix it" to Cursor till it works, and mostly they like that instead of searching StackOverflow and figuring out what to copy from there, this is a much faster approach to coding

3. The senior engineers, neckbeards, old-school nerds wrangling with complexes codebases have mostly been saying "eh... it is good, in places. I don't like the code it writes, but I like how I can ask it to explain API docs and new pieces of code I have not seen before". They are using it to write some pieces of code in languages they are not well versed with, and they prefer 'copilot' mode more than 'chat and generate' mode.

4. The absolute elites (like Linus below), basically just say "it is all hype", and are not impressed at all with it, and are mostly ignoring it.

I think I am somewhere between 2 and 3 myself, and it was quite a bit of cognitive dissonance initially to find my friends from both 2 and 3 buckets give such contradicting signals about it. Then I started seeing the pattern. Those who are objectively worse programmers than me, were objectively more impressed by it than I am. Conversely those who I look up to as much better engineers than me, seem to be impressed even less than me with it.

So yeah, basically LLMs write code like a ~p75 programmer (probably ~p50 when codebase is large), and anyone below that mark is impressed by it, and those above it find it overwhelming, and mostly an irritating pair programmer, whose code they need to keep nitpicking.

[Dipendra Bhatt @dipenbhatt03]

Well i definitely hate being oncall, couldn't even finish my overhead press lest the prod systems catch fire.

[Dipendra Bhatt @dipenbhatt03]

@akaushik759 "aap kamchor hai janab" said not me

[Dipendra Bhatt @dipenbhatt03]

@iNishant You remember back in Dukaan how much emphasis we used to give on this so that Folks build a instinct to do this. That behaviour to always click on the files after raising a PR and just skim through.
I am still preaching this behaviour in my current work lol!

[Ojasvi Yadav @ojasvi_yadav]

looking for Flutter devs for a 0-1 AI app, paid opportunity, bengaluru only
personal references preferred

[Dipendra Bhatt @dipenbhatt03]

@akaushik759 Same boat m hu bro

[Mitchell Hashimoto @mitchellh]

Cloud pricing applied to literally anything else, i.e. the coffee shop: coffee is $5, but if you stay it’s free and we charge $0.06/minute unless you’re using internet then it’s $0.09/minute. First timers get 30 minutes free. Talking to others in the cafe is free. Voice calls outside the cafe are $1/minute unless its to our other cafe locations then its free for the first 12 minutes then $0.50/minute thereafter. If you have any problems write your issue on a napkin and throw it straight into the trash. If you pay for a premium support plan (contact us) then please step into the personalized Lamborghini to talk to your account manager. Enjoy your time at Cloud Cafe, please complete the survey on your way out.

[Dipendra Bhatt @dipenbhatt03]

@yurytom @samokhvalov @lukaseder The managed aspect is really more on the terms of having features like upgrading your db, setting up replica, snapshot, providing observability, etc etc right out of the box.