Samuel Dushimimana Shyaka

Key Skills

Software

Labelbox

Appen

Scale AI

Top Subject Matter

Healthcare - Medical Records & Patient Data

Finance - Risk Analysis & Fraud Detection

E-commerce - Product Categorization & Customer Support

Top Data Types

Image

Video

Text

Top Task Types

Bounding Box

Polygon

Segmentation

Classification

Point Key Point

Object Detection

Freelancer Overview

Machine Learning Engineer (Consultant). Brings 6+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include AWS SageMaker. Education includes Master of Science, Emory University (2025) and Master of Science, Carnegie Mellon University (2024). AI-training focus includes data types such as Text and labeling workflows including Segmentation, Classification, and Fine Tuning.

Labeling Experience

Machine Learning Engineer (Consultant)

Aws SagemakerTextSegmentation

As a Machine Learning Engineer consultant, I built predictive models and data pipelines to improve Pro customer targeting and marketing performance. My work included engineering features from large transactional datasets and deploying interpretable, production-grade pipelines for business impact. Strong skills in XGBoost, PyTorch, BigQuery, and MLOps were essential to excel in this position. • Developed XGBoost models to predict SKU-level repurchase probability and next-month purchase quantity with high accuracy. • Engineered and maintained BigQuery feature pipelines for massive transactional data, generating behavioral insights for 8M+ SKUs. • Built LSTM time-series models to capture complex purchasing patterns for advanced market segmentation. • Designed and deployed batch scoring pipelines with interpretable outputs to improve campaign targeting and conversion rates.

As a Machine Learning Engineer consultant, I built predictive models and data pipelines to improve Pro customer targeting and marketing performance. My work included engineering features from large transactional datasets and deploying interpretable, production-grade pipelines for business impact. Strong skills in XGBoost, PyTorch, BigQuery, and MLOps were essential to excel in this position. • Developed XGBoost models to predict SKU-level repurchase probability and next-month purchase quantity with high accuracy. • Engineered and maintained BigQuery feature pipelines for massive transactional data, generating behavioral insights for 8M+ SKUs. • Built LSTM time-series models to capture complex purchasing patterns for advanced market segmentation. • Designed and deployed batch scoring pipelines with interpretable outputs to improve campaign targeting and conversion rates.

2024 - 2025

Machine Learning Engineer

Aws SagemakerTextSegmentationClassification

In this role, I designed and deployed semantic search and NLP systems for cybersecurity literature analysis. My contributions involved fine-tuning language models, optimizing retrieval performance, and serving real-time semantic query systems. The position required skills in BERT, RoBERTa, PyTorch, FAISS, FastAPI, and cloud model deployment. • Created intelligent semantic search over 3,000 cybersecurity papers, cutting search times by two-thirds. • Fine-tuned transformer models for domain-specific classification and efficient corpus filtering. • Implemented semantic similarity retrieval across 120K+ document segments for improved relevance. • Deployed a real-time FastAPI pipeline serving ranked documents via SageMaker endpoints with low response times.

In this role, I designed and deployed semantic search and NLP systems for cybersecurity literature analysis. My contributions involved fine-tuning language models, optimizing retrieval performance, and serving real-time semantic query systems. The position required skills in BERT, RoBERTa, PyTorch, FAISS, FastAPI, and cloud model deployment. • Created intelligent semantic search over 3,000 cybersecurity papers, cutting search times by two-thirds. • Fine-tuned transformer models for domain-specific classification and efficient corpus filtering. • Implemented semantic similarity retrieval across 120K+ document segments for improved relevance. • Deployed a real-time FastAPI pipeline serving ranked documents via SageMaker endpoints with low response times.

2022 - 2024

Education

Work History