John R. Kosinski

It's a secure system of contracts—one of the first of its kind—that creates an on-chain ETF for US Treasury Bonds and other fixed-income instruments. It is currently launched on the Binance Smart Chain (BSC) and Ethereum. The on-chain side is written in Solidity, off-chain processing is mainly TypeScript on AWS, and the front end is created with vanilla React.

PROJECT LINKS:
• streamprotocol.money
• appstreamprotocol.money/
• etherscan.io/address/0xd86FFB404147163e19E010a0e9d4995E0e36F335
• etherscan.io/address/0xADF789E61Bf38c463e4bA5B2B6E9C1Af6659e11b
https://etherscan.io/address/0xa0b86991c6218b36c1d19d4a2e9eb0ce3606eb48
• etherscan.io/address/0x746430119DA3EA0e398b93059e96F21bfabAa8a4
https://etherscan.io/address/0xD48425B7fb702F571D872f4b7046B30c9FA47e15
• bscscan.com/address/0x3daa26E70383B37E41434a8bF8e8bF3A335c144A
https://bscscan.com/address/0xB95Bc0d1057a43ACF608805c6D4e3e234e4d4C1b
• bscscan.com/address/0x8ac76a51cc950d9822d68b83fe1ad97b32cd580d
https://bscscan.com/address/0x2A7aD6908A08dD19D6c5090aA86E7c7E61C3e7F4
• bscscan.com/address/0x2900E387152a22370F482a0640701e1B41cc0896
• github.com/jrkosinski/evm-bond-vault

Wrote and published an article on the most common design anti-patterns.

Scanning the blockchain's verified NFTs, I frequently see recurring common anti-patterns. I wrote an article listing what I think are the most common ones, some thoughts about why they occur, and some better alternatives.

These are solutions to the well-known Ethernaut problems. Exercises published by OpenZeppelin as teaching tools and to assess one's ability to understand EVM and Solidity security vulnerabilities and how to protect against them. These solutions contain full source code with unit tests and explanations.

This is an article that I wrote after finishing a defi project about a particular design pattern in Solidity to secure a group of smart contracts. It's a simple pattern that can serve as a helpful introduction for individuals to the concept of modular design in Solidity and how it's a bit different than in traditional OOP.

Developed and deployed secure and well-tested NFT smart contract and web3 scripts to manage the contract for EVM chains like Ethereum and Polygon in Solidity and for Solana using Metaplex. Listed on Opensea and Magic Eden.

An award-winning iOS grocery order/delivery app for the well-known NYC online grocery store, FreshDirect. It was developed with .NET using the Xamarin iOS platform. This app won a Webby in 2015 in the websites and mobile sites category.

An iOS app that allows customers of Aquamonix to interact remotely with their industrial-sized SCADA irrigation systems. It was developed from scratch using Xamarin iOS.

This IoT demo for Alexa allows users to control their Samsung TV remote to turn their TV off and on using an Echo device and code embedded in an ESP8266 chip.

It is associated with the Toptal blog article available at toptal.com/arduino/esp8266-arduino-tutorial-alexa-hack

These are my recent Medium articles, showcasing skills and knowledge in
• blockchain development and architecture
• machine learning and AI
• statistics
• software development and architecture
• communication skills

A two-part article detailing and demonstrating some techniques for pre-processing timeseries data (the example used is stock price historical data) in preparation to use it as input to a Keras long short-term memory (LSTM) neural network.

The preprocessing described and demonstrated includes statistical transformations like scaling, handling outliers, removing the trend, and extracting important features. It also includes the correct way to shape the data into the appropriate number of dimensions.

The article is in two parts:
Part 1: https://medium.com/@john.kosinski/preparing-and-shaping-timeseries-data-for-keras-lstm-input-part-one-5bb882bc2143
Part 2: https://medium.com/@john.kosinski/preparing-and-shaping-timeseries-data-for-keras-lstm-input-part-two-ad17f6ab450