Daniel Krastev Email & Phone Number

Arlington, Virginia, Us

New York, Ny, Us

Direct report to the CTO. Led the end-to-end Software Development Lifecycle of several machine learning codebases in Python. Acquired a keen awareness of business strategy by virtue of abundant participation in board-level business meetings.--------------------1) Semantic Search (Lead Developer) – Won a multi-million dollar contract with the Department of Defense's (DoD) Defense Innovation Unit (DIU). Led the R&D effort of a semantic search engine that processes a corpus of 1M+ documents of unstructured text data. It is capable of ingesting high-throughput streaming data, and performing unsupervised, real-time, semantic disambiguation, token level processing, parsing, and named entity recognition (NER). Developed the full backend in Python, built a government-compliant REST API using Flask, configured a government-compliant authentication layer, and performed cloud deployment on AWS.2) Topic Deltas – Surfaces earnings calls insights for thousands of companies. Re-engineered existing codebase, from prototype, to production level.3) Rumor Hound (Lead Developer) – Researched and developed a novel machine learning algorithm which evaluates a digital news-source’s reliability in publishing valid, market-moving M&A chatter. Engineered a contextually- adaptive web crawler interfaced with MongoDB. Featured a dynamically- learning price-prediction model, which evaluates a company's potential price move and relative market impact, benchmarked to the S&P 500.

San Diego, California, Us

Developed accuracy and performance tests for Teradata’s analytics functions within Python-driven automated testing frameworks. Extensive experience with Agile Software Development.--------------------1) Patent Pending - “Teradata Advanced Analytics Platform and Tensorflow for Character Recognition”. Developed a scalable machine learning architecture by integrating Tensorflow, SQL-MapReduce (SQLMR), and Hadoop Distributed File System (HDFS).2) Developed an autocorrect algorithm in Java, which was successfully implemented in a Massively Parallel Processing architecture and integrated within a big data pipeline. Featured as a demo at Teradata’s annual Partners Conference.3) Developed and executed mathematically rigorous quality tests on “Teradata’s Generalized Linear Model with Elastic Net, Lasso, and Ridge Regularization” machine learning function.4) Developed synthetic data sets for algorithmic research & development.

Princeton, Nj, Us

Designed an underactuated mechanism that can emulate the biomechanical characteristics of a human finger.--------------------1) Implemented cross-cutting graduate-level mathematics and engineering concepts to enable passively-adapting finger-flexion.2) Design demonstrated levels of functionality consistent with market products at less than 1/10th of the cost.

Bachelor Of Science In Engineering (B.S.E), Mechanical And Aerospace Engineering

High School