Skip to content
View kovashikawa's full-sized avatar
🚧
building
🚧
building

Block or report kovashikawa

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
kovashikawa/README.md

about me

I'm Rafael, a Data Scientist and AI Engineer specializing in scalable data systems, forecasting, NLP, and applied machine learning. I design end-to-end data and AI solutions, from high-performance pipelines to production-ready LLM and RAG applications.

My experience spans hedge-fund macro research, real-time data platforms, credit risk systems, and evaluation frameworks for time-series forecasting models. I am completing the MIT Statistics and Data Science MicroMasters and previously studied Economics at FGV-EPGE.

I work across the full stack of data science and engineering: modeling, infrastructure, APIs, MLOps, LLM systems, and quantitative analysis.


skills

core technical stack

  • Python, R, SQL, MongoDB
  • Airflow, AWS, GCP, Kubernetes, FastAPI, Polars
  • Machine Learning, NLP, LLMs, RAG architectures, Econometrics
  • Tableau, Plotly, Matplotlib
  • GitHub (CI/CD), Linux, Docker

languages

  • English
  • Portuguese
  • Spanish

selected experience and projects

  • Quantitative macro modeling, CPI forecasting systems, and NLP pipelines for monetary policy at a global hedge fund (5+ Billion USD AUM)
  • Large-scale ETL/ELT redesign, reducing infrastructure cost by half
  • High-performance simulation API for credit risk using FastAPI and Polars
  • Production RAG system for legal document interpretation using LangChain
  • Evaluation frameworks and simulation environments for advanced time-series forecasting models (including LLM-based forecasters)

See public projects in this profile, including micro-services, LLM tools, and ML pipelines.


outreach and training

  • 2023: Machine Learning using TensorFlow (USP)
  • 2020: Quantitative Finance with Statistical Analysis of Financial Data in R (FGV)
  • 2019: Fundamentals of Marketing Analytics with Machine Learning (FGV)
  • 2018: Python project using PCA, K-means, DBSCAN for security analysis (FGV)

awards

  • Big Data Hackathon (XP Inc. & Microsoft Azure, 2021): built a big data pipeline and analysis workflow on Azure using >14GB of raw data
  • Cryptocurrency Datathon (FGV & Ripple, 2020): second place for ML-driven crypto trading and NLP-based market signal extraction Video presentation: https://youtu.be/_aCNF3jHSss?t=671 Repository: https://github.com/kojabawa/pirates

Let's keep in touch! visit my LinkedIn profile: https://www.linkedin.com/in/rkovashikawa/

Pinned Loading

  1. spotify2youtube spotify2youtube Public

    Transfer playlists from spotify to youtube

    Python 1