





This is from Ganpati visarjan in Pune. these are very young school kids, and honestly, this was just painful to watch. The kind of gestures they were making, along with wearing footwear on the rath, felt completely inappropriate for the occasion. ganpati celebrations have always been such an important part of Pune’s culture, so seeing things like this is genuinely disappointing. it does make me wonder what we are teaching the next generation.
The latest #Android 17 QPR1 update to Pixels seems to have pushed the changes concealing the developer and debugging state from 3rd party apps so,
finally the developers can use the damn banking and other apps that have these security checks!
it seems to be live on, Pixel 6A running CP3A.260905.009
And confirm it has not yet landed on the beta channel.
A friend sent me this 4.4 MB BIN file on WhatsApp, I found it suspicious and didn't open it. I called him, and he said his WhatsApp was hacked.
I'm curious how this actually works. How do attackers compromise a WhatsApp account and then send malicious files to the victim's contacts? The scary part is that people are much more likely to trust a file when it comes from someone they know.
How does this attack work, and what's the best way for a normal user to protect themselves?
I wanted to share some thoughts on why I really appreciate what DeepSeek is doing in the AI space.
What stands out to me is that DeepSeek seems to treat AI/LLM technology primarily as an engineering and research challenge. They are focused on making models better, more efficient, and more accessible rather than constantly creating noise around AGI timelines, predictions, or fear-driven marketing.
A lot of major US AI labs are doing incredible work, and their models are impressive. But sometimes the progress feels heavily centered around bigger training runs, benchmark improvements, and larger models. Again, those achievements are valuable, but the progress is often measured mostly through benchmark numbers.
With DeepSeek, every release feels like it comes with some actual engineering innovation.
Their open-weight approach and research papers are already a huge positive. But beyond just releasing a model, they keep introducing interesting improvements, things like architectural optimizations, efficiency improvements, and techniques such as KV cache reduction. They are not just saying "here is a better benchmark score"; they are showing how to make AI systems more efficient.
The inference side is another thing that amazes me. Building an efficient inference stack is not easy, and seeing the kind of performance DeepSeek achieves with extremely competitive pricing is genuinely impressive. The ability to provide such capable models while keeping inference costs low shows a lot of engineering excellence.
Another thing I like is their communication style. Their social media presence is refreshingly different. There is no constant hype, no unnecessary announcements, no marketing drama. They release their work, publish the research, and let developers and users judge the results.
And the pricing is another strong point. Providing powerful models at such competitive prices while maintaining sustainability is honestly impressive. It shows that efficiency and optimization can matter just as much as simply scaling compute.
For me, DeepSeek represents a very engineering-focused philosophy of AI development: fewer words, more research; less hype, more innovation.
Keep building, DeepSeek. Looking forward to seeing what comes next.
Ultimately, a person should not ask what the meaning of life is. Instead, they must realize that life is asking the same question to them. Each person is questioned by life, and they can only answer-not with words-but by taking responsibility for their own life and the way they choose to live it.
Hey everyone, I recently graduated in CS engineering and I'm currently working as a technical consultant at a fintech MNC. Pay is decent, no complaints there, but I've been thinking a lot about where I actually want to be in the next five to ten years and honestly I'm stuck.
One option is doing a Masters after gaining a year or two of experience, probably in the US. But financially this feels like a huge bet for my family. I'd have to leave my job, and the kind of money needed would most likely mean taking on debt. If anything goes wrong along the way, it could really hurt us financially, and that risk keeps holding me back even though part of me wants to do it.
Second option is an MBA back home. Same problem of having to leave my job for close to two years, but the bigger issue for me is that an MBA usually pushes people toward management roles. And honestly I don't want that. I like being on the technical side and I don't want to lose that edge just to move into people management.
Third option, which honestly feels the most natural right now, is just focusing on prepping well and switching to a better role. Strengthen my skills, target better companies, and grow that way. But even here I keep thinking there's a ceiling to this approach. At some point as an individual contributor you become expensive for a company to justify keeping around, and if you want to keep climbing after that point, it feels like moving into management becomes almost unavoidable.
So yeah that's my dilemma. I want to keep growing technically but I also don't want to box myself into a corner ten years down the line. Would really appreciate hearing from people who've actually navigated this, especially anyone who managed to stay technical while still growing in seniority and pay. What would you do in my position, and what do you wish someone had told you earlier. Thanks for reading through all this.
Tweet Made by Official Handle of Maharashtra Congress.
I'm a new joiner at Oracle and had a quick question about the enterprise access we get to ChatGPT and Codex. I know these are company-provided tools, but I'm curious about how much visibility the company actually has into our usage.
Can managers or IT/admin teams see the actual prompts or chat history, or is it limited to high-level usage metrics like active users, number of chats, etc.? I'm just trying to understand how it works from a privacy and administration perspective before I start using it more regularly.
Would appreciate any insights from people who've looked into this. Thanks!
Oracle laid off 500 employees in Romania with emails titled 'COLLECTIVE DISMISSAL - YOU ARE A PROPOSED IMPACTED EMPLOYEE.' This marks the second major workforce cut there in less than a year, echoing similar layoffs in the US and India.
Link to Article
I'm a recent graduate and have started working as a consultant. I want to keep learning outside of work and prepare myself for future software engineering roles.
One thing I've never really figured out is how people actually study from technical books such as DDIA, System Design Interview, etc.
Do you just read them, take notes, highlight sections, make summaries, or something else entirely? I always feel like I'll forget everything if I don't take notes, but when I try taking detailed notes my reading becomes extremely slow.
For those who have learned a lot from technical books, what does your actual process look like?
I just started working as a consultant, pay is decent but it’s not the core engineering work I want long term. So I’m trying to keep learning on the side to prep for future SWE roles.
I’ve got the usual books, Designing Data-Intensive Applications, System Design Interview, and a few others. My problem is I don’t know how to actually use them. If I just read, I forget most of it in a week. If I highlight everything it feels pointless. If I take full notes, one chapter takes me three evenings and I never finish.
For people who’ve actually gotten through these, what worked for you? Do you do handwritten notes, short digital summaries, flashcards, build a tiny project after each chapter, or just re-read?
Would really appreciate your real process. Thanks.
I'm a recent CSE graduate from a Tier-2 college with a 9+ CGPA, and I'll be joining a large MNC next month as an Consultant through campus placements. The package is around 9 LPA, and honestly, I'm excited because I've always enjoyed working in tech.
At the same time, I've been thinking a lot about my long-term career path. One option I'm seriously considering is working for around 1.5-2 years, gaining some industry experience, and then preparing for CAT with the goal of getting into a top MBA college like the IIMs, FMS, and similar institutes.
The reason I'm confused is that I genuinely enjoy the technical side of things. I like building things, learning new technologies, working on engineering problems, and staying close to the product and development side. Pursuing an MBA from a top institute sounds attractive because of the opportunities, network, career growth, and compensation potential, but it also feels like I might gradually move away from the technical work that I currently enjoy.
At the same time, I've seen many people in tech grow significantly through skill development, experience, and job switches without necessarily pursuing an MBA. Because of that, I'm trying to understand whether a top MBA would actually create opportunities that are substantially different or better than what I could achieve by continuing on the technical path.
I also wanted to ask about CAT preparation itself. If I decide to pursue this route after gaining some work experience, is it realistic to prepare through self-study while working full-time? Have any of you managed to score well and convert top colleges without joining expensive coaching programs?
Looking forward to your insights. Thanks!
I have a budget of $5 to $10 to top up my API balance. Since both models now cost the same, I wanted to ask those who have used both: which one is the better option?