Rahim Khoja

"Trader" is a cutting-edge cryptocurrency trading platform developed by Improbability Labs Inc., designed to streamline and enhance the trading experience for users of all skill levels through sophisticated algorithmic trading. It features the Technical Analysis Trade Processor (TA-Bot), which leverages customizable algorithms and a range of technical indicators to help users optimize their trading strategies based on historical data. Unique to "Trader" is its automated portfolio management… Show more

"Trader" is a cutting-edge cryptocurrency trading platform developed by Improbability Labs Inc., designed to streamline and enhance the trading experience for users of all skill levels through sophisticated algorithmic trading. It features the Technical Analysis Trade Processor (TA-Bot), which leverages customizable algorithms and a range of technical indicators to help users optimize their trading strategies based on historical data. Unique to "Trader" is its automated portfolio management system, which intelligently adjusts trading strategies to maximize performance and mitigate risks. The platform supports extensive back-testing and live trading functionalities, providing users with the tools to test, refine, and execute strategies effectively. With its user-friendly interface and powerful technical architecture, "Trader" is set to redefine investment strategies in the digital age, offering a proactive and dynamic trading environment that promises superior financial outcomes.

• Led the design and development of 'Trader', an advanced algorithmic trading and portfolio management platform targeting cryptocurrency and stock markets.

• Architected robust system infrastructure using Slurm HPC, Proxmox/KVM, and Linux, focusing on scalability and high availability.

• Engineered sophisticated data analysis tools and trading strategies using Python, integrating libraries such as Pandas, TA-Lib, and Numpy.

• Implemented robust Flask APIs and utilized Grafana for system monitoring, significantly enhancing operational efficiency and insight.

• Integrated key financial data APIs and pioneered the use of AI with OpenAI's GPT models for predictive analytics, improving trade decision-making. Show less

A 1.4 Petabyte sized Chia Cryptocurrency farm. Runs on 8 Dell Servers with a ZFS storage backend. All plots created with my customized Bash/SystemD script that managed multiple drives.

I had the privilege of developing the backend for Compass, an AI-powered chatbot designed to deliver legal and regulatory insights tailored for the cryptocurrency domain. This innovative system, part of the Crypto-Co-Counsel project, connects users with accurate and up-to-date information through a private and secure chat interface. The backend leverages advanced Language Learning Models (LLMs) and Retrieval-Augmented Generation (RAG) to provide real-time, data-driven responses.

The… Show more

I had the privilege of developing the backend for Compass, an AI-powered chatbot designed to deliver legal and regulatory insights tailored for the cryptocurrency domain. This innovative system, part of the Crypto-Co-Counsel project, connects users with accurate and up-to-date information through a private and secure chat interface. The backend leverages advanced Language Learning Models (LLMs) and Retrieval-Augmented Generation (RAG) to provide real-time, data-driven responses.

The project also integrates Ollama to deploy models on private instances, ensuring privacy and secure data handling alongside utilizing AI providers like OpenAI and Anthropic. The frontend for this project was expertly developed by Graham Morley, enabling a user-friendly interface for seamless interaction.

Key Features:
🔹 Multi-LLM Integration: Works with leading AI providers (OpenAI, Anthropic, Ollama, GroqCloud) while also supporting privately hosted models for enhanced privacy.
🔹 RAG for Credible Responses: Retrieves curated legal documents dynamically, enhancing responses with inline references to ensure accuracy and transparency.
🔹 Data Residency Compliance: Implements user-specific PostgreSQL databases on Azure to meet region-based residency requirements.
🔹 Real-Time Interaction: Enables low-latency communication using SocketIO for fast and reliable chat experiences.
🔹 Scalable Architecture: Deployed via Fly.io with Dockerized infrastructure to ensure high availability and global scalability.
🔹 Security by Design: Incorporates JWT-based authentication, SSL encryption, and SOC 2-level compliance to protect user data and interactions.

This project allowed me to build a robust system that simplifies access to complex legal information for cryptocurrency stakeholders, balancing cutting-edge technology with user privacy and compliance. For more details, feel free to reach out or visit https://www.cryptococounsel.com/. Show less

📈 "The Quant" is a powerful AI-driven stock analysis tool designed to empower traders by predicting potential trading opportunities using artificial intelligence. Hosted on a Vultr Ubuntu VM, this tool is tailored for enthusiasts who appreciate data-driven decision making.

How It Works:
1. Input your OpenAI API Key: This allows OpenAI's models to process and analyze your stock data.
2. Select an AI Model: Choose between GPT-4o or GPT-4 Turbo for cutting-edge analysis.
3… Show more

📈 "The Quant" is a powerful AI-driven stock analysis tool designed to empower traders by predicting potential trading opportunities using artificial intelligence. Hosted on a Vultr Ubuntu VM, this tool is tailored for enthusiasts who appreciate data-driven decision making.

How It Works:
1. Input your OpenAI API Key: This allows OpenAI's models to process and analyze your stock data.
2. Select an AI Model: Choose between GPT-4o or GPT-4 Turbo for cutting-edge analysis.
3. Provide the Stock Ticker: Enter the ticker symbol of the stock you're interested in.
4. Choose Your Analysis Period: Opt for Daily, Weekly, or Monthly analysis to suit your trading strategy.

🔍 Our tool fetches OHLCV data, applies a comprehensive set of technical indicators, and passes the processed data to the AI. Based on this, it provides actionable insights, whether you're looking to buy, hold, or short a stock. The AI recommends optimal trading strategies, including Buy at Market, Limit Sell orders for profit-taking, and Trailing Stop Losses for risk management.

🛠️ Bring Your Own API Key: To maintain transparency and control, users need to bring their own OpenAI API Key to ensure that all analyses are personalized and secure.

📊 Whether you're a seasoned trader or just starting, "The Quant" equips you with the insights needed to navigate the complexities of stock markets with confidence. Show less

This unique subscription service is designed to enhance your daily productivity by performing Google lookups directly via SMS. Whether you're in Canada or the USA, you can now access information on the go, simply by texting your queries!

Key Features:
- Local SMS Numbers: Tailored for a more personal user experience.
- Personalized AI Assistant: Customized interactions to suit your needs.
- Multilingual Response Capability: Engage in your preferred language.
- Real-Time… Show more

This unique subscription service is designed to enhance your daily productivity by performing Google lookups directly via SMS. Whether you're in Canada or the USA, you can now access information on the go, simply by texting your queries!

Key Features:
- Local SMS Numbers: Tailored for a more personal user experience.
- Personalized AI Assistant: Customized interactions to suit your needs.
- Multilingual Response Capability: Engage in your preferred language.
- Real-Time Data Retrieval: Powered by Google for the latest information.
- Rapid Development: Concept to launch in just 5 weeks, thanks to collaborative efforts with ChatGPT and Llama 3 from Groq!

A huge shoutout to Kelly Williamson for his incredible support and expertise throughout the development process. https://lnkd.in/g3DKhd3R- Developed the SMS AI Assistant, a comprehensive AI-driven platform designed to interact with users via SMS, leveraging Flask, MySQL, NGINX, and CertBOT.

- Integrated with OpenAI and GroqCloud to provide intelligent conversational capabilities, enhancing user engagement and service value.

- Incorporated payment processing through Stripe and communication services via Twilio and SendGrid, ensuring a seamless user experience.

- Utilized Ubuntu 23 on Vultr hosting for deployment, ensuring robustness and reliability of the service. Show less

A docker image with Jupyterlab that includes Sage Math and many more Plugins and Addons. Build from the data science notebook but massively updated and customized.

Detect Crypto Currency buying opportunities from Technical Indicators.

Dynamic DNS Server is a web application and API that enables users to manage virtual DNS hosts. The server is compatible with most DDNS clients as its API imitates the dyndns2 protocol. This open-source project, was developed as a personal endeavor to become proficient with Tomcat, Bind, and CentOS 7.

To accomplish this project, the following tasks were completed:

- Configured, compiled, and deployed Bind 9.1x with DLZ support on a CentOS 7 server.

- Installed and… Show more

Dynamic DNS Server is a web application and API that enables users to manage virtual DNS hosts. The server is compatible with most DDNS clients as its API imitates the dyndns2 protocol. This open-source project, was developed as a personal endeavor to become proficient with Tomcat, Bind, and CentOS 7.

To accomplish this project, the following tasks were completed:

- Configured, compiled, and deployed Bind 9.1x with DLZ support on a CentOS 7 server.

- Installed and configured MySQL 5.7 as both a master and a slave. Slave mode only replicated specific tables.

- Configured and deployed Tomcat 9 on CentOS 7, with credentials obtained from the database.

- Developed a simple CRUD-based web application on Tomcat 9 for DNS record, domain, user, and other Dynamic DNS functionality management. Users can create new hosts, delete hosts, update host IPs, and retrieve access to their accounts. Admins can add new domains, delete domains, and manage users.

- Designed a straightforward relational database for the web application and integrated it with the Bind DLZ database schema for easy management.

- Developed a Tomcat-based REST API that emulates the dyndns2 protocol. A JDBC realm that authenticates against the database was used to secure the API, which was tested against several DDNS clients for compatibility.

- Created several systemd service files, conf files, bash scripts, and cron jobs.

- Designed the system to utilize Slave DNS servers for production. Slave DNS Servers are Bind with DLZ and MySQL in slave mode installed on CentOS 7. Show less

As a personal project, I developed the Atomik Controller, which was later turned into a product to help attract investors to cover the certification costs of Mi-Light Wireless Bulbs for North American Markets. The Atomik Controller is a versatile device that can control over 100,000 different wireless RGB bulbs. It was built on a Raspberry Pi with an Nrf24l01+ 2.4 GHz transceiver and written in C++11, node.js, PHP, MySQL, JQuery & Bash. It leveraged various technologies such as Linux (Debian)… Show more

As a personal project, I developed the Atomik Controller, which was later turned into a product to help attract investors to cover the certification costs of Mi-Light Wireless Bulbs for North American Markets. The Atomik Controller is a versatile device that can control over 100,000 different wireless RGB bulbs. It was built on a Raspberry Pi with an Nrf24l01+ 2.4 GHz transceiver and written in C++11, node.js, PHP, MySQL, JQuery & Bash. It leveraged various technologies such as Linux (Debian), Cron, Nginx, and Systemd.

To bring the Atomik Controller to fruition, I researched the certification requirements for electrical and transmitting products in Canada and the USA, collaborating with the Canadian Standards Association and the FCC. I also reverse-engineered the custom Mi-Light on-air protocol, as well as their 802.11 smartphone protocol, to enhance the initial controller design by increasing the total number of addressable devices from 4 to 130,560.

As part of the project, I designed a Node.JS JSON API that enables third parties to integrate with and control the Atomik Controller with ease. Additionally, I built a Mi-Light smartphone server emulator that was secured with Mac address filtering. I utilized Cron to create timers that enable light settings to change for different light zones, and designed a web GUI to manage all the light bulbs, RF remotes, smartphone remotes, system network settings, scheduled tasks, and other system settings.

Please go to the following Hackaday.io and Github links for more information:
https://hackaday.io/project/12748-the-atomik-controller Show less