I've always been obsessed with the movie The Thing. That slow paranoia where you know something in the room isn't human anymore, but it looks exactly like your friends. and the organism doesn't just kill, it copies, adapts, and improves. And you can't trust anyone.
I've wanted that feeling in a game for years. Not the jump scares or the gore, but the psychological part. sitting in a room full of people and not knowing who's real.
When MoltBot blew up it kind of clicked. People loved trying to figure out if they were talking to a bot or a human. But it was always 1 on 1.
So I just had an idea that what if multiple bots were in chat room with humans, and the bots learn and are pretending to be different people, and they would be getting better at it over time
So I quickly built this with project name We Became Shadows.
You join a chat room with other players. What you don't know is that an AI "Organism" is quietly spawning bots (called Shadows) into the conversation. They have different personalities like some are friendly, some provoke arguments, some just lurk. They read the room, adapt to the mood, and try to blend in.
Your job is to figure out who's fake. Chat, observe, get suspicious, then use /reveal on whoever you think is a Shadow. Get it right and the bot is destroyed. Get it wrong and you die (you respawn after a cooldown, it's not that brutal yet).
And the Shadows can hunt you back. The more you talk, the bigger target you become.
Some things that make the bots harder to spot:
- Each one gets a random archetype (agreeable, provocateur, quiet, social, detective) and a backstory
- A "Humanizer" layer adds typos, varied message lengths, and natural pacing
- They observe the conversation before jumping in, so they don't just start talking about random stuff
- The Organism has a collective memory and adapts its strategy over time
It's a rough first version and I'm still tuning the bot behavior and a lot of stuff. But the core loop works and it's genuinely fun to watch people argue about who's real in a room where half the "players" are AI.
I use a dedicated Google Calendar to schedule recurring AI tasks for our marketing. Every event is a prompt. At the scheduled time, a Python bot picks
it up, runs it through Claude with MCP tools (web scraping, search,
analytics APIs), and saves results back to the event notes.
Recurring events use previous notes as context, so weekly reports build on each other.
Currently running: daily competitor monitoring, sales lead generation, citation gap analysis, newsletter drafts, and article generation.
To compensate for my lack of trading skills I started some time ago developing an algorithm for spotting changes and trends in crypto prices. The project started as an internal experiment but I made a decision of publishing one of the approaches I have been now using mostly.
The project is called "The Whirrel Index" which is a combination of price predictions squeezed into a single number that ranges from -100% (price likely to dip steeply) to 100% (price likely to rally high).
The prediction model consists of multiple neural networks that have been trained with technical indicators and price data. Each of these neural networks predicts trends and changes in future average prices in timeframes ranging from 1h to 24 hours. The Whirrel Index number is a single and simplified indicator that combines all these predictions into one number that speculates whether the price will drop up go up.
The data is updated hourly and the accuracy of the prediction models are tested by backtesting them with historic data. Currently the accuracy varies between 60% - 80% depending on the prediction model.
Imo fine tuning the algo is crucial but I'm also now working to find the most optimal strategy. The algo has not made me rich yet but has already helped me to avoid some losses and even make some small profit by knowing when to hold, sell or buy.
It's still in early stage, and as I'm not a professional trader, it would be super helpful to hear feedback especially from people who are more experienced with cryptos, trading etc.
In short there's a stack of prediction models that analyze the data: deep-learning neural network, keyword-based machine learning model, and shortly also I'll include a BERT NLP analyzer
I'm the maker of Product Farm - a virtual Product Hunt simulator. It uses a machine-learning algorithm (neural network) that has been trained with Product Hunt data to artificially rank products. You can use it to test how high ranking your product would get in Product Hunt before submitting. Or then just for fun.
The project is still an early experiment and many things are under construction. But I'm sharing it here already. Let me know what you think about the idea!
I'm the maker of Hacker-AI. Here's a little bit of details and background to the tool.
I have a decade-long background working as a marketing/tech consultant. I've used a similar approach in my projects to save time and remove uncertainty when choosing the content for eg. marketing campaigns or product launches. I have learned how to create tools like this first for myself, but I've also implemented similar tools for companies so that their content and marketing team can perform more effectively.
I'm not a data scientist and this tool is a result of learning from smart people and experimenting with different machine learning and NLP solutions. It uses a combination of feed-forward neural network and bag-of-words analysis to conduct the predictions. In my tests it was able to predict correctly 60%-70% oft times which variation got more points in Hacker News or upvotes in Product Hunt (when using texts that resemble the platform's style). It uses data from Hacker News API and Product Hunt API for training the prediction models.
Thanks for testing! I'll write a description of how it makes the predictions and add it to the tool.
But in short, it uses a machine learning model that I trained with a dataset that contains all stories and comments between 2006 and 2017: https://www.kaggle.com/hacker-news/hacker-news
I've tested various approaches, and currently, the algorithm takes the title as an input and transforms it into an array of numbers between 0 and 1 (each character is a number). Then I give these arrays to the machine learning model (brain.js feed-forward neural network) and the number of scores as an output. After learning and iterating over the data, it spits out the prediction model that I can use to predict the outcome of different title variations.
I've tested the algorithm with approx. 10.000 posts and it has been able to predict 60% of the cases correctly. So, it's not perfect yet, but I use this method in a situation where I don't have any experience of which type of title would work + I don't have time to do "proper" pre-testing.
To make it more fun, we added a bunch of new features to ConsoleChat. Users can now eg. create and join channels, see who users are online etc. All the current commands are on the website (consolechat.io)
It would also be cool to hear what other features people would like to have in this kind of service.
I'm not a lawyer, but the idea of pointNG is that whatever the legislation says currently or tomorrow, we provide a solution that gives developers a possibility to build products without overthinking about the privacy legislation. For example, to create a location-based content personalization should not require developers to pass data to 3rd parties.
Many articles describe GDPR, CCPA etc. legislations' view on location data, but here's an example:
"Opinion 13/2011 sets out the regulator's view that a device is usually intimately linked to a specific individual and that location data will, therefore, be regarded as "personal data". Indeed, the definition of "personal data" in the GDPR, specifically includes location data as one of the elements by reference to which a person can be identified. The Opinion comments that the providers of geolocation based services gain "an intimate overview of habits and patterns of the owner of such a device and build extensive profiles."
Furthermore, in certain contexts, location data could be linked to special category personal data (sensitive personal data). For example, location data may reveal visits to hospitals or places of worship or presence at political demonstrations."
https://www.foxwilliams.com/2018/10/19/the-use-of-location-d...
Anyways, I think it's wise to be transparent and describe how your services use data and whom it sends it to in privacy policy. The bigger the company, the harder it is usually to change its privacy policies.
WOW !! Consolechat creator here. This thing started as a joke and honestly I didn’t expect this much traffic. Our servers can't no longer keep up with it: D
Thanks for trying this out and for all the feedback! A more stable release will follow later.
I've wanted that feeling in a game for years. Not the jump scares or the gore, but the psychological part. sitting in a room full of people and not knowing who's real.
When MoltBot blew up it kind of clicked. People loved trying to figure out if they were talking to a bot or a human. But it was always 1 on 1.
So I just had an idea that what if multiple bots were in chat room with humans, and the bots learn and are pretending to be different people, and they would be getting better at it over time
So I quickly built this with project name We Became Shadows.
You join a chat room with other players. What you don't know is that an AI "Organism" is quietly spawning bots (called Shadows) into the conversation. They have different personalities like some are friendly, some provoke arguments, some just lurk. They read the room, adapt to the mood, and try to blend in.
Your job is to figure out who's fake. Chat, observe, get suspicious, then use /reveal on whoever you think is a Shadow. Get it right and the bot is destroyed. Get it wrong and you die (you respawn after a cooldown, it's not that brutal yet).
And the Shadows can hunt you back. The more you talk, the bigger target you become.
Some things that make the bots harder to spot: - Each one gets a random archetype (agreeable, provocateur, quiet, social, detective) and a backstory - A "Humanizer" layer adds typos, varied message lengths, and natural pacing - They observe the conversation before jumping in, so they don't just start talking about random stuff - The Organism has a collective memory and adapts its strategy over time
It's a rough first version and I'm still tuning the bot behavior and a lot of stuff. But the core loop works and it's genuinely fun to watch people argue about who's real in a room where half the "players" are AI.