Buenos Aires, Provincia de Buenos Aires, Argentina
642 seguidores
Más de 500 os
Acerca de
Actividad
-
Finally! After over 2.5 years we published our Digital Product Portfolio Management guide as of today: https://lnkd.in/gk9g4GKG I wanted to thank…
Finally! After over 2.5 years we published our Digital Product Portfolio Management guide as of today: https://lnkd.in/gk9g4GKG I wanted to thank…
Recomendado por Matias Selser
-
Python is not a perfect language. It has some gotchas that are confusing to newbies but make sense once you learn the object model. The problem is…
Python is not a perfect language. It has some gotchas that are confusing to newbies but make sense once you learn the object model. The problem is…
Recomendado por Matias Selser
Experiencia y educación
Licencias y certificaciones
Publicaciones
-
Optimal Market Making by Reinforcement Learning
MACI 2021 - pp. 471 - 475
We apply Reinforcement Learning algorithms to solve the classic quantitative finance Market Making problem, in which an agent provides liquidity to the market by placing buy and sell orders while maximizing a utility function. The optimal agent has to find a delicate balance between the price risk of her inventory and the profits obtained by capturing the bid-ask spread. We design an environment with a reward function that determines an order relation between policies equivalent to the original…
We apply Reinforcement Learning algorithms to solve the classic quantitative finance Market Making problem, in which an agent provides liquidity to the market by placing buy and sell orders while maximizing a utility function. The optimal agent has to find a delicate balance between the price risk of her inventory and the profits obtained by capturing the bid-ask spread. We design an environment with a reward function that determines an order relation between policies equivalent to the original utility function. When comparing our agents with the optimal solution and a benchmark symmetric agent, we find that the Deep Q-Learning algorithm manages to recover the optimal agent. We apply Reinforcement Learning algorithms to solve the classic quantitative finance Market Making problem, in which an agent provides liquidity to the market by placing buy and sell orders while maximizing a utility function. The optimal agent has to find a delicate balance between the price risk of her inventory and the profits obtained by capturing the bid-ask spread. We design an environment with a reward function that determines an order relation between policies equivalent to the original utility function. When comparing our agents with the optimal solution and a benchmark symmetric agent, we find that the Deep Q-Learning algorithm manages to recover the optimal agent.
Otros autoresVer publicación
Proyectos
-
Chainlink Hackaton - Squinf
- actualidad
As a developer on the Squinf project, I was responsible for modifying the architecture of Opyn's Squeeth smart contract to create a new inflation tracker token. This involved working with multiple tokens as collateral and deg tests using historical data to measure the performance of the new token. I also designed an oracle to connect to an inflation market data source, ensuring that the token remained up to date with the latest market trends.
Achievements:
- Successfully…As a developer on the Squinf project, I was responsible for modifying the architecture of Opyn's Squeeth smart contract to create a new inflation tracker token. This involved working with multiple tokens as collateral and deg tests using historical data to measure the performance of the new token. I also designed an oracle to connect to an inflation market data source, ensuring that the token remained up to date with the latest market trends.
Achievements:
- Successfully modified the smart contract architecture to create a new inflation tracker token.
- Designed tests using historical data to measure the performance of the new token.
- Created an oracle to connect to an inflation market data source, ensuring the token remained up to date with the latest market trends.
- Contributed to the development of a project that has the potential to revolutionize the way inflation is tracked and managed in the financial industry.
Through my work on Squinf, I have gained valuable experience in developing smart contracts, deg tests to measure performance, and working with market data sources. I am confident that the skills and experience I have gained will be valuable in any future projects I undertake in the blockchain and financial industries. -
Bitstamp Bot
- actualidad
As a software developer, I designed and implemented a high frequency trading bot that connected to Bitstamp using FIX protocol to real-time data and deploy trading algorithms. This project involved the use of Python and C++ programming languages, as well as various technologies such as MongoDB, Flask, and React.
Key achievements:
- Designed the generic architecture of the high frequency trading bot, which included the use of FIX protocol to connect to Bitstamp and…As a software developer, I designed and implemented a high frequency trading bot that connected to Bitstamp using FIX protocol to real-time data and deploy trading algorithms. This project involved the use of Python and C++ programming languages, as well as various technologies such as MongoDB, Flask, and React.
Key achievements:
- Designed the generic architecture of the high frequency trading bot, which included the use of FIX protocol to connect to Bitstamp and real-time data.
- Created the FIX protocol market connectors for market data and order flow for both Python and C++ programming languages. This allowed for seamless integration of the trading bot with the Bitstamp trading platform.
- Designed and implemented a front-end interface that displayed real-time tick-by-tick data for a given set of trading pairs. The interface included a menu that allowed s to define metrics and boundaries for their trading strategies.
- Used MongoDB to store and manage trading data, which enabled efficient retrieval and analysis of historical trading data.
Overall, this project allowed me to develop my skills in software development, data management, and trading strategy design. I gained valuable experience in using FIX protocol to connect to trading platforms, deg and implementing front-end interfaces, and managing large amounts of trading data. -
Fantom Arbitrage Bot
-
As the lead developer of a high frequency trading bot, I designed and implemented a multi-dex and multi-wallet ed bot to arbitrage tens of thousands of currency pairs in real-time using Go and smart contracts. In this role, my achievements include:
- Designed the mathematical model of the arbitrage strategy under the blockchain ecosystem, taking into the complex variables involved in high frequency trading on decentralized exchanges.
- Implemented arbitrage pairs…As the lead developer of a high frequency trading bot, I designed and implemented a multi-dex and multi-wallet ed bot to arbitrage tens of thousands of currency pairs in real-time using Go and smart contracts. In this role, my achievements include:
- Designed the mathematical model of the arbitrage strategy under the blockchain ecosystem, taking into the complex variables involved in high frequency trading on decentralized exchanges.
- Implemented arbitrage pairs finding algorithms using graph theory, enabling the bot to quickly and efficiently identify profitable trading opportunities across multiple exchanges.
- Developed the multithreading high frequency trading bot backend in Golang, enabling the bot to execute trades at a high frequency with minimal latency.
- Deployed the database and trading bot in EC2 AWS instance, ensuring scalability and reliability of the bot's performance.
- Created and implemented a smart contract to perform the swaps between multiple AMMs in the blockchain, enabling seamless and secure trading across different decentralized exchanges.
Overall, my work on this project demonstrates my expertise in deg and implementing complex trading systems using cutting-edge technologies. I am excited to bring my skills and experience to future projects in the field of blockchain and high frequency trading.
Reconocimientos y premios
-
Optimal Market Making by Reinforcement Learning
-
I was extended an invitation to participate in a meeting with the trading desk of Crypto Finance Group in Zurich to engage in a comprehensive discussion about the paper titled "Optimal Market Making by Reinforcement Learning."
The purpose of the meeting is to delve deeper into the paper's findings and explore potential applications of the reinforcement learning approach to market making strategies.
The meeting presented a unique opportunity to collaborate with experienced…I was extended an invitation to participate in a meeting with the trading desk of Crypto Finance Group in Zurich to engage in a comprehensive discussion about the paper titled "Optimal Market Making by Reinforcement Learning."
The purpose of the meeting is to delve deeper into the paper's findings and explore potential applications of the reinforcement learning approach to market making strategies.
The meeting presented a unique opportunity to collaborate with experienced professionals in the finance industry and gain valuable insights into the implementation of cutting-edge technologies in the trading world.
Idiomas
-
English
Competencia bilingüe o nativa
-
Italian
Competencia básica
-
Spanish
Competencia bilingüe o nativa
-
Portuguese
Competencia básica
Más actividad de Matias
-
¡Sumate y conversá con nuestros graduados! 🎓 ¿Estás considerando hacer la Maestría en Finanzas Di Tella? Conocé de primera mano sus experiencias a…
¡Sumate y conversá con nuestros graduados! 🎓 ¿Estás considerando hacer la Maestría en Finanzas Di Tella? Conocé de primera mano sus experiencias a…
Recomendado por Matias Selser