Ask HN: What is your company doing with AI?
What is the company that you work for doing with AI or trying to do with AI? Just trying to get a more realistic sense of how AI will affect society and the workplace.
13 comments
As a caveat, my job isn’t programming but I am self-taught in Python and Google Apps Script.
ChatGPT has been incredible for the small marketing agency where I work to do much more w/ automations. Being able to use natural language to describe a script that will work with the Asana API (for example) has been amazing. There’s been projects that would’ve taken me hours in the past that are now done in minutes.
There’s also been a lot of data collection and research we’ve been doing by having ChatGPT write small scripts — like scrape this CSV of URLs and return all the meta titles/descriptions. There were tools for these things in the past but the flexibility of easily scripting them ourselves has been great.
In the past, the vast majority of this stuff would’ve either not gotten done or would’ve been half-assed by me — maybe sometimes I’d have hired a freelancer to help. But now I just say to keep the automation ideas coming because my time is no longer the limiting factor.
ChatGPT has been incredible for the small marketing agency where I work to do much more w/ automations. Being able to use natural language to describe a script that will work with the Asana API (for example) has been amazing. There’s been projects that would’ve taken me hours in the past that are now done in minutes.
There’s also been a lot of data collection and research we’ve been doing by having ChatGPT write small scripts — like scrape this CSV of URLs and return all the meta titles/descriptions. There were tools for these things in the past but the flexibility of easily scripting them ourselves has been great.
In the past, the vast majority of this stuff would’ve either not gotten done or would’ve been half-assed by me — maybe sometimes I’d have hired a freelancer to help. But now I just say to keep the automation ideas coming because my time is no longer the limiting factor.
Nothing, with no plans to. I don't technically work for a company though, but a government entity serving a rural community and not about extracting money from people. A static website and email/phone/in-person communication works well for what we're doing.
I run a gamedev studio and make a lot of titles in Godot. ChatGPT is an absolute beast at GD Script 3.5.3 and also parsing scene files - I assume because Godot is entirely open source with lots of docs online. It will occasionally hallucinate built-in functions but that's easily caught and rectified. Saved me a ton of time when I had to write custom pathfinding scripts. It's also given us more velocity with features where we can kind of just toss off fun "juice" ideas to it (like minigames or easter egg type content) and get back a workable script in minutes that would've been like a weekend stretch goal.
A pet project I have in my head is using another LLM with a huge context window and being able to dump in all the scene and script files so the AI has insight into the entirety of the game and how the nodes interact rather than just the isolated scripts I send. Not sure if this is feasible locally with the same level of accuracy as Claude or ChatGPT
A pet project I have in my head is using another LLM with a huge context window and being able to dump in all the scene and script files so the AI has insight into the entirety of the game and how the nodes interact rather than just the isolated scripts I send. Not sure if this is feasible locally with the same level of accuracy as Claude or ChatGPT
At the company I work for, we're leveraging AI to enhance our product's capabilities in voice recognition, subtitle translation, and generation.
-Our platform uses advanced AI models to deliver highly accurate transcriptions, which is essential for creating precise subtitles. -The AI-driven translation feature ensures that subtitles are not only accurate but also contextually appropriate, making them more natural and easier to understand. -We've designed the user interface to be straightforward and user-friendly, allowing even those without technical expertise to easily navigate and use our tools.
Our product is set to launch soon, so stay tuned!
-Our platform uses advanced AI models to deliver highly accurate transcriptions, which is essential for creating precise subtitles. -The AI-driven translation feature ensures that subtitles are not only accurate but also contextually appropriate, making them more natural and easier to understand. -We've designed the user interface to be straightforward and user-friendly, allowing even those without technical expertise to easily navigate and use our tools.
Our product is set to launch soon, so stay tuned!
We're a developer platform [0], for running apps at scale and locally.
The platform is batteries included: global router, logs, metrics, managed resources (datbases, caches, block storage). Because we have all this data in a unified platform we can (already, without AI) correlate issues across different system components, reliably and without elaborate integration work. We're now taking that a step further and using LLMs to fulfill a portion of the support/operational role.
As it happens, we're hosting an event on these developments for NY Tech Week [1].
0. https://noop.dev
1. https://lu.ma/h015tm6c
The platform is batteries included: global router, logs, metrics, managed resources (datbases, caches, block storage). Because we have all this data in a unified platform we can (already, without AI) correlate issues across different system components, reliably and without elaborate integration work. We're now taking that a step further and using LLMs to fulfill a portion of the support/operational role.
As it happens, we're hosting an event on these developments for NY Tech Week [1].
0. https://noop.dev
1. https://lu.ma/h015tm6c
We have several processes which involve the upload of multiple different images, which the user is walked through with baby steps, I.e. 'take this exact picture of the asset' -> take the next, etc. Etc.
We're currently working on an image classification model to automatically tag these images with what they are and just providing a visual sample gallery of 'take these pictures' with the sample image border turning green when that type of image is identified. We expect this will reduce a multi-step error prone process with a more streamline experience.
We're currently working on an image classification model to automatically tag these images with what they are and just providing a visual sample gallery of 'take these pictures' with the sample image border turning green when that type of image is identified. We expect this will reduce a multi-step error prone process with a more streamline experience.
I'm assuming by "AI" you mean the new crop of LLMs here. At my company, we've been using various forms of "AI", including deep learning systems, for many years in our products and are continuing to do so.
The devs here have tried out the new LLM stuff, as well as things like copilot, but to my knowledge nobody has incorporated them into their workflows because they've found no advantage to doing so.
The devs here have tried out the new LLM stuff, as well as things like copilot, but to my knowledge nobody has incorporated them into their workflows because they've found no advantage to doing so.
We use:
- LLMs to make support team more efficient
- Github Copilot to make writing code easier
- A slew of SaaS products that themselves are using LLMs to deliver current and future features
- LLMs to make support team more efficient
- Github Copilot to make writing code easier
- A slew of SaaS products that themselves are using LLMs to deliver current and future features
There will be very few companies which will be really using the AI to solve problem which AI can solve, otherwise every next company slapping "AI" on their prodcut.
We use lots of AI, in the sense that machine learning algorithms are AI.
... where I've worked before:
sorry, but my managers were to stupid for even MS excel. I hope, i will never ever again meet such people. If it ever happens though, I will leave the next day. You cant imagine what a struggle it is - even more of a struggle, when the premise is "manage up" and not "team up".
The best part: Don't even understand what security is and means, but being really proud to announce taking part in security workshop for software architects. They don't even work in that area. The company did not even engage in any kind of code writing. The only code I've written there was large, nested formulas into the excel. And then, you have one, your manager - who is proud to take part in a security workshop???? ... They didn't even understand what that means. I suddenly lost all of my respect the second I read that in a mail conversation.
The same for the AI Workshop, where one university wanted to discuss how they want to manage chatgpt usage with minors and home assignments. You can't believe it - "I take AI seriously" was the fastest answer I've ever heard.
That will stay this way. But, its up to me now :) My plan is to build their processes around AI and by this put them out of business where it hurts the most. Its too easy, if you've worked there for 14 years and being the only one who have done everything from A to Z. Being "the actual company" ...
So my plan is to utilize LLM and some image interpretation techniques. It is a highly technical thing. I can overview the business and what is happening there and I can tell, it will put some people out of work. If I really do this. By now, just trying to come up with a way of self-financing. Then to get my co-workers out of my ex-employers claws (they also can use my self-financing methods) and then we'll start with that :)
so, of course it will affect society and workplace.
sorry, but my managers were to stupid for even MS excel. I hope, i will never ever again meet such people. If it ever happens though, I will leave the next day. You cant imagine what a struggle it is - even more of a struggle, when the premise is "manage up" and not "team up".
The best part: Don't even understand what security is and means, but being really proud to announce taking part in security workshop for software architects. They don't even work in that area. The company did not even engage in any kind of code writing. The only code I've written there was large, nested formulas into the excel. And then, you have one, your manager - who is proud to take part in a security workshop???? ... They didn't even understand what that means. I suddenly lost all of my respect the second I read that in a mail conversation.
The same for the AI Workshop, where one university wanted to discuss how they want to manage chatgpt usage with minors and home assignments. You can't believe it - "I take AI seriously" was the fastest answer I've ever heard.
That will stay this way. But, its up to me now :) My plan is to build their processes around AI and by this put them out of business where it hurts the most. Its too easy, if you've worked there for 14 years and being the only one who have done everything from A to Z. Being "the actual company" ...
So my plan is to utilize LLM and some image interpretation techniques. It is a highly technical thing. I can overview the business and what is happening there and I can tell, it will put some people out of work. If I really do this. By now, just trying to come up with a way of self-financing. Then to get my co-workers out of my ex-employers claws (they also can use my self-financing methods) and then we'll start with that :)
so, of course it will affect society and workplace.
Building prototypes for clients to explore genAI
mainly just as a copilot to do repetitive work for you