warthog

this is amazing, great idea

really like these - curious how the xlsx editor and viewer is built in that what kind of headless spreadsheet?

could not have been easy

website has no self serve sign up

GPT-5.5 on the benchmarks still seem to perform better than this

Plus the vibe of the gemini models are so weird particularly when it comes to tool calling

At this point I kinda need them to shock me to make the switch

Hoping the benchmarks are correct this time...

is this something like sprites?

vertical tabs are a great idea for ghostty!

https://banker.so/ - Retool alternative through vibecoding

- Building a Retool/Looker alternative for internal tools so non-technical teams can build their own without creating busywork for engineers. - Works like Lovable/Replit with two differences: Got 3,000 integrations out of the box through Pipedream SDK and got access controls in place with embedded auth, backend (django) and database

Analysis done with Claude Code and then Codex to verify. Each agent was given API keys in free tiers to play around and test the sandboxes for various things.

https://mertdev.dev/

Can you explain how you achieve this in more detail? Did not see any in-detail explanation in the readme in repo

Had a pretty bad year. Hoping 2026 will be way better! Happy new year everybody

There was an old legendary HN comment around the following concept:

"All things being equal, people buy from their friends. So just make more friends"

If you are confident of your services, just make more friends. Talk to people more. If it is too salesly already, you lost the conversation. Instead try and learn about their lives. I sometimes even open up with "No agenda to sell, genuinely curious"

Note: If anyone finds the link to the comment, please drop it here to credit the author.

Traditionally the time of the year for Spotify Wrapped variations! Coded with Claude Code.

Run `uvx claude-code-wrapped` or `npx claude-code-wrapped`

Tried a couple times but it failed

Banker.so | Computer inside a computer inside an agent

Started this out by building a spreadsheet controlled by an LLM. Now putting a direct filesystem inside, simplified enough to have programmatic control of slide builders, spreadsheets, terminals and vibecoding applications

what would be the difference of this vs using an API like Apollo?

This looks awesome. Although from a UX perspective might not be as good as streaming token by token for text generation use cases. However for image gen and editing - 100%

Built an in-browser AI managed spreadsheet. https://banker.so/

Too many things I wanted to analyse went to nothing because I was too lazy to fetch information and the put it inside a spreadsheet cell by cell. So I developed this to help me extract docs into spreadsheets while also having access to the web.

Now working on a vibe-coded version of it where instead of showing a spreadsheet, it will be able to generate data-focused tiles and apps

Actually just wrote about this: https://aimode.substack.com/p/openai-is-below-above-and-arou...

not sure if it binary like that but as startups we will probably collect the scraps leftover indeed instead

Tough day to be an AI Excel add-in startup

yep hit me up

sorry to hear that. curious if the product's approach was a fuck you to Workday though as in you can't even put them side by side and compare (they are so different) or if it was simply that Workday sucks, we will do better

Paul Graham also has a good way to frame this which perhaps I should have touched upon.

"A principle for taking advantage of thresholds has to include a test to ensure the game is worth playing. Here's one that does: if you come across something that's mediocre yet still popular, it could be a good idea to replace it. For example, if a company makes a product that people dislike yet still buy, then presumably they'd buy a better alternative if you made one."

I definitely wrote it by hand, no LLM used if that is what you are insinuating.

Uber might have turned to the bad practices used by taxis now that they are focused on extracting more and more value. However, the point of the writing was to be focused on earlier days. In the earlier days, they did try and embodied price transparency and customer experience focused on customers.

If you are arguing that they never did, I don't see how it grew to a $200 Bn company.

what do you mean by regulations? Those that benefitted Uber or?

imo it would be better to carry the whole memory outside of the inference time where you could use an LLM as a judge to track the output of the chat and the prompts submitted

it would sort of work like grammarly itself and you can use it to metaprompt

i find all the memory tooling, even native ones on claude and chatgpt to be too intrusive

i think it is too difficult to manage with little upside. setting it up is easy but managing the nodes etc is a huge overhead.

what is the difference of this to conductor

I built this using OpenAI GPT-5 agentic loops as well where the agent interacts with a spreadsheet: banker dot so

Some principles I learned: 1. You don't really wanna use any kind of framework that wraps around openai or claude sdks. You end up fighting them 2. Small number of tools with more functions are better than large number of tools with small number of functions inside 3. Function definitions you put inside the tools are actually very important and need to be very clear 4. Integrating another agent as a tool to the main agent can relieve your context window by 5x 5. RAG via only vector search is mostly unnecessary. Claude Code itself leverages iterative search which works way better with code repos as well as most documents