Aidyn Kadyr

Own and operate 13 production GCP data-pipeline services ingesting multi-source company data into a unified canonical PostgreSQL schema for downstream analytics and ML. Built serverless BigQuery data lake pipelines provisioned via modular Terraform, including Companies House bulk ingestion with partitioned vintage query tables and an LLM enrichment pipeline with JSON-schema constrained outputs.

Developed a Python algorithm to consolidate historical quote data for 1,200+ customers, improving retrieval by 10x and translating requirements for product and marketing decision-making. Built time-series demand-forecasting models on historical sales data to support supply-chain feasibility, risk management, and inventory planning.

Developed an internal revenue-forecasting model in Palantir Foundry using Python and PySpark, collaborating in an Agile environment to align technical work with financial decision-making. Implemented TDD practices to raise test coverage from 30% to 85% and ensured dataset synchronization from the DWH into Palantir.

Developed XGBoost classification and regression models for trading illiquid forex pairs, prototyping algorithmic MVPs and continuous performance-monitoring systems. Performed time-series analysis and backtesting to guide feature selection, achieving 32% simulated ARR across a 10-year historical backtest.

Pursuing a Master of Science in Data Science and Business Analytics at Bocconi University, developing advanced skills for data-driven analytics and ML.

Completed a Bachelor of Science in Mathematics at Nazarbayev University, building a strong foundation in quantitative methods.