Udit Khandelwal (@uditkhandelwal8) 's Twitter Profile
Udit Khandelwal

@uditkhandelwal8

Founder @AppRunner
| Simplifying VM app deployment for indie hackers & vibe coders. Ex-Microsoft/Amazon, NIT Warangal. Build easy! bit.ly/4lVuErr

ID: 1415688330

linkhttps://apprunner.synergiqai.com calendar_today09-05-2013 14:47:50

461 Tweet

71 Followers

101 Following

Andrew Ng (@andrewyng) 's Twitter Profile Photo

Happy 2026! Will this be the year we finally achieve AGI? I’d like to propose a new version of the Turing Test, which I’ll call the Turing-AGI Test, to see if we’ve achieved this. I’ll explain in a moment why having a new test is important. The public thinks achieving AGI means

Garry Tan (@garrytan) 's Twitter Profile Photo

Starting a company is probably the only way to try to escape bullshit jobs: reporting, compliance theater, internal dashboards, meetings to coordinate meetings, process to manage fear, maintaining status hierarchies, liability shields, and managerial control, not producing value.

Udit Khandelwal (@uditkhandelwal8) 's Twitter Profile Photo

Of late claude code's $20 plan has become non useful for coding. In the middle of one coding session, the limit expires in an hour. Time to move to $100 plan ? #ClaudeCode

Udit Khandelwal (@uditkhandelwal8) 's Twitter Profile Photo

OpenAI has come a long way in code generation and is doing a better job than Claude Sonnet. While Sonnet often overthinks, makes mistakes, and requires hand-holding, OpenAI's GPT-5.2 with medium reasoning performs much better. It also handles more context and delivers stronger

Udit Khandelwal (@uditkhandelwal8) 's Twitter Profile Photo

Wow! He literally spelled out the entire hacking process step by step. Of course, it only works for one use case and the steps aren't applicable to all, but the process is real. It's problem-solving under the hood.

Udit Khandelwal (@uditkhandelwal8) 's Twitter Profile Photo

The message is clear. Tune website as per the new standards. The faster you do it, the better your chances are of winning in this AI race.

Udit Khandelwal (@uditkhandelwal8) 's Twitter Profile Photo

I understand that LLMs are expensive but in Claude code pro, the reduced limits are making it unusable. Claude soon people may drift away and look for alternates. May be its time to make haiku default and improve on the agent if sonnet is too expensive.

Udit Khandelwal (@uditkhandelwal8) 's Twitter Profile Photo

Not sure if goose is the answer but we do need local AI model for coding to make it scalable and that local model needs to learn from the coder coding it. #vibecoding

Sumeet Chawla (@sumeetchawla) 's Twitter Profile Photo

A Capterra study found that 72% Indian consumers used AI chatbots for product search and intend to keep using them, and 51% said they prefer AI chatbots for online shopping. The way brands are discovered is changing fast. seerly.app/blog/why-ai-se…

A Capterra study found that 72% Indian consumers used AI chatbots for product search and intend to keep using them, and 51% said they prefer AI chatbots for online shopping.

The way brands are discovered is changing fast.

seerly.app/blog/why-ai-se…
Rakesh Menon (@rakeshmenon13) 's Twitter Profile Photo

Folks using Openclaw: Optimise the agent’s SOUL, HEARTBEAT and MEMORY retrieval instructions. Agents are not autonomous or have special abilities. Your instructions and your model drive it. Beware of the compactions. Mark important instructions as CRITICAL or NON-NEGOTIABLE.

Udit Khandelwal (@uditkhandelwal8) 's Twitter Profile Photo

Its like a catch 22 situation. How do we build payment history in the next 90 days if a major payment instrument is blocked out due to lack of it.