img

MLH fellowship and others stopped for Indians.

Source https://www.reddit.com/r/developersIndia/comments/1e38hx2/no_internationational_programs_for_indian_students/

img
img

satan

Stealth

2 months ago

img

Elon_Musk

X.com

2 months ago

img

satan

Stealth

2 months ago

See more comments
img

FreshDew5

Oracle

2 months ago

img

FreshDew5

Oracle

2 months ago

img

tbk

Startup

2 months ago

See more comments
img

Noctus

Stealth

2 months ago

img

jake_peralta_B99

Unemployed

2 months ago

img

tbk

Startup

2 months ago

img

jake_peralta_B99

Unemployed

2 months ago

img

dr_AR

Student

2 months ago

img

tbk

Startup

2 months ago

img

dr_AR

Student

2 months ago

img

BT7274

SAP

2 months ago

Sign in to a Grapevine account for the full experience.

Discover More

Curated from across

img

Software Engineers on

by PushyCourt

Others

Indian SDEs will do all DSA and System Design chatter but cannot create any real value

Indian SDEs on LinkedIn & Twitter are some of the worst engineers you will ever find across the real world. Their typical value lies around: - Cracking a FAANG/MAANG or Product company. - Make an announcement on LinkedIn & Twitter and accumulating followers. - Starting YouTube channel, Topmate links, affiliate marketing channels, and more. - Rant DSA, System Design, Interview resources all day long. - Post daily of their office, their colleagues, candid shots, and barely getting any work done. While this is expected out of early career professionals, it pains to see senior SDEs doing the same shit over and over again. Things don't end here — Everyone talks about creating a revolution. And the revolution is teaching DSA. If you are an experienced SDE and still ranting DSA, it just means that you have barely learnt anything that is of worth any value. You are atmost a "Ticket Engineer". You get assigned a Jira ticket, you bust your ass solving it, and that's it. That's your entire value. Its high time that these "Bhaiya", "Didis", "Bhabhis" stop ranting DSA and System Design over and over, and build something of real value. You would barely see any credible open-source projects coming out of India (but hey, we have some many contributors!), barely any Indie dev project, or anything that captures the attention of the world. Literally every popular SDE you might follow or see over your social media is the same. Their entire community game is a farce and designed to capture gullible college students and shill thousands of rupees out of them. You might be featured on Times Square ($40 ka showoff), but no one would remember you for anything that you built. <Rant Over. Peace>

img

Product Managers on

by BahadurBilla

Stealth

Cordially inviting all deserving tech bros

In July 2002, Google did something crazy: They hired a 22 year old computer science student and made him the product manager for Gmail. Gmail engineers were shocked. Gmail was the most important new project at Google. Its target was 10M users. And this guy, who was just out of college, would be the PM? But they were overruled, because Google had a bigger problem: Product managers. Google couldn’t find good PMs. Co-founder Larry Page rejected all the super-experienced PMs from Microsoft, McKinsey, etc. They talked about management and strategy and business. Larry Page hated this. He wanted Google PMs to be technical. Otherwise how would they work with the engineers? That’s when a Google VP, Marissa Mayer, had an idea: If we want PMs with technical skills and don’t care about their experience, why don’t we hire computer science students from college? Thus, the Associate Product Manager (APM) program was born. Brian Rakowski, a 22-year-old CS grad from Stanford, became the first APM. He was put on the Gmail team. Brian was scared. How could he work with super experienced, super senior engineers on a super important project if they didn’t respect him? Marissa gave him the answer: Data She explained to Brian that a PM didn’t give orders to engineers. His job was to be helpful to engineers. If he had an idea, first he should do a small test on 1% of users. If it worked, he should show the test data to the engineers to work on it. Thus, data became the centre of decisions, not seniority or politics. The rest is history. Gmail was a huge success. APM program was a huge success. The first APM, Brian, was a huge success. He’s currently the Vice President of Product Management at Google! Reference : https://www.linkedin.com/posts/shreyans512_in-july-2002-google-did-something-crazy-activity-7192496184895115264-nKXF?utm_source=share&utm_medium=member_desktop