While many cryptographic components of blockchain protocols can beextremely complex, blockchain systems themselves are relatively easy tounderstand when viewed from a distance. To take the example of Bitcoin,users store digital currency in hardware or software wallets, and useprivate keys to sign and broadcast transactions. Broadcastedtransactions are grouped together into a block through a cryptographicprocess known as mining, with miners rewarded through the collection oftransaction fees and the issuance of new coins. The mined block oftransactions is appended to the existing chain, and verified by a globalnetwork of nodes. This process repeats in perpetuity, with each newlyadded block adding to the trustedness and security of data stored on thechain.Increased interest in and demand for cryptocurrencies has brought abouta need for places where digital assets can easily be bought, sold ortraded. Our platform, Bitpanda, accomplishes this with a backend writtenin Python, and relying heavily on Django and MySQL databases. In ourpresentation, we begin by providing a brief overview of how blockchainswork. Following this, we describe the Python architecture that (e.g.)generates cryptocurrency wallets, builds, signs and sends transactions,and monitors blockchains for new, relevant data. Key challenges,solutions and failures encountered during the development of the system,and growth of our team, are presented.Throughout our talk, we also highlight a number of broader socialimplications of blockchains, and our work with them. More specifically,we describe the need for open-innovation based approaches to blockchaindevelopment, the value of open-source within the blockchain community,and the current lack of critical discourse surrounding the potentialuses of blockchains as mechanisms of surveillance and control.