Subham Singh

Building @tokenguardrail

[Subham Singh @subhamsng]

Launched on @ProductHunt

ClientKit: Client Management & GST Invoicing for Freelancers https://t.co/BwSiOffQtQ via @producthunt

#buildinpublic #saas #crm #clientkit #freelancer #freelance #agency

[Subham Singh @subhamsng]

@jonegptbc https://t.co/Uw1It1qily

[Subham Singh @subhamsng]

@borrhensaidi @X I tried this. Loved the idea 🚀

[Subham Singh @subhamsng]

@joseph_mugen Personal Brand

[Subham Singh @subhamsng]

I think @vercel’s analytics is good enough. I would be using it for all my projects instead of GA.

#analyticstool #WebsiteDevelopment #seo #vercel

[Subham Singh @subhamsng]

@danellisona Should we build without any customer or interest shown on the product? @danellisona

[Akshay Saini @akshaymarch7]

Gen Z has a low attention span, and we can't blame them for it.

Millennials and Boomers are slowly getting even worse.

Stop expecting every new generation to change for your comfort. Accept them as they are. Try to understand them.

Don't curse what you don't understand.

They might be different, but they're not always wrong.

[Subham Singh @subhamsng]

Hi @dearsummer0, I’m @tokenguardrail which helps to save you from overspending on LLM API calls. You can wrap your LLM calls with our SDK and set budget rules on the dashboard which basically caps all the API calls and send alert when the limit reaches instead of letting it burn more and more money.

[Subham Singh @subhamsng]

To everyone building AI agents or apps - how are you budgeting your AI usage right now?
I'm building something to solve exactly this. If it sounds interesting, let's connect.

#buildinpublic #saas #aiagents #anthropic #claude #openai #softwarecompany

[Subham Singh @subhamsng]

@piyushgarg_dev @piyushgarg_dev I believe some sort of monitoring on the usage and budgeting should be there when we're using those Pay as go APIs for our AI apps/agents.
I'm building a tool for solving exactly this, would love you feedback on the idea Piyush.
please checkout @tokenguardrail

[Subham Singh @subhamsng]

@i_mika_el It cuts the stream off at the source, so you stop paying the instant the budget's hit, not after.
Your app gets a clean signal to handle it however you want: show a limit message, fall back to a cheaper model, whatever fits.

[Subham Singh @subhamsng]

we have SDK which you can install and wrap all your LLM calls (we don't see any content or API keys).
After your LLM calls are wrapped, you can setup the budgets on the dashboard and we'll enforce the usage based on the rule set on the dashboard.
We've made sure that everything is compliant with the policies globally.
Do check out : https://t.co/6g1Ssxx5QB for more.

[Subham Singh @subhamsng]

@newworldkhaled hahah yes

[Subham Singh @subhamsng]

@hthieblot 😈 heheh

[Subham Singh @subhamsng]

@danielkleach a tool which saves you from overspending on LLMs
https://t.co/MnviSMvIcd

[Subham Singh @subhamsng]

@AnandaniNisha 6 😭 only 2 are active

[Subham Singh @subhamsng]

@hthieblot Save LLM costs

[Akash Bhadange @designerdada]

We are probably experiencing the best and the worst of the internet right now.

[Subham Singh @subhamsng]

Agreed!
FYI, @master_jpma I'm building a tool that wraps your LLM API calls and lets you set budgets per agent, project, or user, so if an agent loops out of control, you get an alert instead of a burned budget.
Please signup for the waitlist if you're interested.
https://t.co/6g1Ssxx5QB

[Subham Singh @subhamsng]

What do you do while all the agents are doing the actual work?

#claudecode

[Subham Singh @subhamsng]

@cinamarina Awesome