Mutsa
AI Data Automation Specialist I Dashboards, Python, SQL, Streamlit, R
Habilidades
Revisa mis servicios


Porfolio
Experiencia laboral
Data Scientist
City of Bayswater • Tiempo parcial
May 2025 - Present • 1 yr
• Built an end-to-end Python/R pipeline for high-frequency wastewater sensor data (1-min resolution, 50k+ observations), enabling scalable data accumulation for time-series modeling. • Engineered lagged and rolling multivariate features (N H3, H2S, flow), capturing delayed system dynamics and reducing chemical dosing variance by 18%. • Applied event-study methodology to operational transitions (HCl, FeCl3), estimating causal effects of dosing changes on odor emissions. • Developed quantitative response metrics (baseline shift, % change, persistence, time-to-min) to evaluate treatment strategies, improving performance by 12%. • Developing predictive and optimization models to learn system response dynamics and enable data-driven control of chemical dosing under uncertainty.
Full Stack Data Engineer
Croatia Autentica • Tiempo parcial
Jan 2025 - May 2026 • 1 yr 4 mos
• Architected and deployed a SQL-backed membership, scheduling, and training management platform (Streamlit + MySQL) for 400+ users, operating as a forward-deployed engineering system with real-time feedback loops and continuous feature iteration. • Built end-to-end data pipelines for bookings, fleet usage, and training logs, enabling real-time utilization tracking and automated anomaly detection in scheduling workflows. • Developed interactive calendar-based scheduling and logging systems with constraint enforcement (availability, crew assignment, role-based access), improving resource allocation and operational efficiency. • Designed analytics dashboards (Plotly + SQL aggregations) for membership, financial health, and fleet utilization, supporting data-driven decision-making for club operations.
Junior Engineer
Coca-Cola • Tiempo completo
Jan 2024 - Dec 2024 • 11 mos
• Led data-driven root-cause analyses (RCA) on production line failures, reducing material waste by 9% and improving process efficiency by double-digit margins. • Analyzed machine downtime patterns (MTTR, MTBF) and proposed digitized logging frameworks, improving data availability and enabling predictive maintenance workflows. • Supported implementation of preventive and condition-based maintenance (CBM) strategies across bottling lines, increasing equipment uptime by 12%. • Standardized equipment data collection and contributed to maintenance scheduling across critical systems (labellers, shrink wrappers, compressors, boilers), improving reliability and operational visibility.