🧠 Licensed Professional Engineer with over 25 years of experience designing municipal water and wastewater infrastructure.

Throughout my career, I've worked on projects that help communities move, treat, and manage one of their most critical resources. Those experiences taught me that good engineering is rarely about perfect information. It's about making sound decisions with the best information available.

Today, I'm combining that engineering experience with Python, software engineering, and applied machine learning to build tools that help utilities make better infrastructure decisions.

My interests span infrastructure analytics, predictive maintenance, engineering software, probabilistic modeling, and decision-support systems. While each repository explores a different problem, they all contribute toward the same long-term vision:

Using software and analytics to improve how infrastructure is designed, managed, and maintained.

I believe machine learning creates the greatest value when it becomes part of reliable engineering software that practitioners can trust, not simply another notebook experiment.

Throughout my engineering career, I've seen firsthand how critical decisions are often made with incomplete data, disconnected systems, and manual workflows.

While modern engineering has embraced advanced modeling and digital design tools, there is still tremendous opportunity to improve how infrastructure data is analyzed, interpreted, and transformed into actionable insights.

This GitHub documents my journey toward that goal.

Each repository explores a different aspect of infrastructure analytics—from hydraulic engineering software and cost intelligence to predictive maintenance, anomaly detection, and applied machine learning. Individually, they solve specific engineering problems. Together, they represent a broader vision for applying modern software engineering and data science to the infrastructure industry.

My objective isn't simply to build machine learning models or backend applications.

It's to develop practical engineering tools that help utilities, consultants, and infrastructure owners make more informed, data-driven decisions.

As this portfolio continues to grow, I hope it demonstrates not only what I've built, but where I believe infrastructure engineering is heading.

My work is centered around applying software engineering and data-driven techniques to solve real-world infrastructure problems.

Current areas of focus include:

• 🏗️ Infrastructure Analytics – Asset risk modeling, predictive maintenance, and engineering decision-support systems
• 🧠 Applied Machine Learning – Risk modeling, anomaly detection, probabilistic modeling, time-series analysis, and forecasting
• 💧 Engineering Software – Building practical software tools for water and wastewater engineering
• 📊 Data Engineering – ETL pipelines, data quality, reproducible workflows, and ML-ready datasets
• ⚙️ Backend Development – Designing reliable APIs and services that operationalize engineering and analytical workflows

Engineering has always been about making informed decisions with imperfect information.

Whether designing a water distribution system, evaluating a lift station, or planning future infrastructure investments, engineers rarely have complete data. Success comes from combining technical knowledge, engineering judgment, and the best information available at the time.

That philosophy guides every project in this portfolio.

I don't view machine learning as a replacement for engineering expertise. Instead, I see it as another engineering tool that can help identify patterns, quantify uncertainty, and support better decision-making.

Likewise, software engineering is more than writing code. Well-designed software makes engineering knowledge repeatable, scalable, and accessible to others.

My goal is to build practical engineering software that transforms data into information and information into better infrastructure decisions.

The projects below represent my primary areas of focus and collectively support my long-term vision for Infrastructure Analytics. While each project addresses a different engineering challenge, they all share a common objective: applying modern software engineering and data-driven techniques to improve how infrastructure is designed, managed, and maintained.

Each project is being developed with an emphasis on engineering rigor, maintainable software architecture, and practical real-world applications.

Toronto Pipeline Failure Predict

A real-world applied machine learning project focused on predicting water main failure risk using public infrastructure, environmental, and asset data from the City of Toronto. The project is intentionally designed as the foundation for a deployable infrastructure analytics platform, not a standalone model.

This work prioritizes data realism, engineering constraints, and defensible modeling decisions, reflecting the realities faced by municipal utilities where data is incomplete, heterogeneous, and operationally constrained.

Status: 🟢 Active Development Primary Competencies: Applied Machine Learning • Infrastructure Analytics • Feature Engineering • Risk Modeling • Data Engineering

• Asset-level feature engineering (installation year, material, diameter, environmental exposure)
• Spatial and temporal aggregation of infrastructure and failure data
• Failure labeling under incomplete and inconsistent municipal records
• Model evaluation aligned with risk prioritization, not academic metrics
• Explicit handling of uncertainty and data limitations

• Backend service to operationalize risk scoring and model inference
• Secure authentication and role-based access for internal and stakeholder users
• API endpoints for asset-level risk queries and scenario analysis
• Dashboard integration for infrastructure risk visualization (Power BI or equivalent BI tooling)
• Deployment-focused design emphasizing reproducibility, auditability, and model versioning

• ML treated as a system component, not a notebook artifact
• Clear separation between data processing, modeling logic, and serving layers
• Design decisions driven by how results will be consumed by engineers, planners, and decision-makers

Goal

Deliver an applied ML–driven infrastructure risk system that bridges modeling, backend services, and operational dashboards—demonstrating end-to-end thinking from raw municipal data to deployable decision support.

Long-Term Vision

Develop a production-ready infrastructure analytics platform that enables utilities to prioritize asset replacement, quantify infrastructure risk, and support engineering decision-making through applied machine learning and modern software engineering.

Hydraulic Analysis Suite

A Python-based hydraulic engineering platform designed to support the analysis and design of water and wastewater infrastructure. The project combines established hydraulic principles with modern software engineering practices to improve engineering workflows, promote reusable calculations, and provide a foundation for future infrastructure analytics applications.

Rather than functioning as a collection of standalone calculators, the Hydraulic Analysis Suite is being developed as a modular engineering toolkit where calculations, engineering objects, and design utilities can be reused across multiple projects and applications.

Status: 🟢 Active Development

Primary Competencies: Engineering Software • Hydraulic Analysis • Object-Oriented Design • Python • Software Architecture

• Object-oriented engineering models for hydraulic components
• Friction loss calculations using industry-standard equations
• Total Dynamic Head (TDH) calculations
• Engineering validation and input checking
• Comprehensive unit testing
• Modular project architecture for future expansion

• Pump performance analysis
• System curve generation
• Hydraulic profile calculations
• Gravity sewer hydraulics
• Water distribution utilities
• Engineering reporting
• Interactive visualization tools

• Translating engineering calculations into reusable software components
• Applying modern software engineering principles to traditional engineering workflows
• Building maintainable, extensible engineering libraries rather than isolated calculation scripts
• Emphasizing testing, validation, and reproducibility for engineering applications

Goal

Develop an open, extensible hydraulic engineering platform that serves as a foundation for engineering analysis, education, infrastructure analytics, and future decision-support applications.

Suspicious Sensor

A probabilistic infrastructure analytics project focused on investigating anomalous pressure sensor behavior within municipal water distribution systems. The project combines statistical modeling, anomaly detection, data quality assessment, and engineering judgment to distinguish sensor faults from legitimate system events.

Rather than treating every anomalous reading as a system failure, this project explores how uncertainty can be quantified and incorporated into engineering decision-making.

Status: 🟢 Active Development

Primary Competencies: Infrastructure Analytics • Statistical Modeling • Anomaly Detection • Probabilistic Analysis • Data Engineering

• Synthetic infrastructure dataset development
• Data validation and quality assessment
• Exploratory data analysis
• Statistical characterization of pressure behavior
• Engineering-focused anomaly investigation
• Reproducible analytical workflows

• Probabilistic anomaly detection
• Bayesian inference for event classification
• Time-series analysis
• Multi-sensor correlation
• Risk scoring
• Interactive engineering dashboards

• Combining engineering judgment with statistical analysis
• Distinguishing instrumentation faults from legitimate hydraulic events
• Evaluating uncertainty rather than relying on deterministic thresholds
• Developing analytical workflows suitable for real-world utility operations

Goal

Develop an infrastructure analytics platform capable of assisting utilities in identifying, prioritizing, and investigating abnormal system behavior while reducing false alarms through probabilistic modeling and engineering context.

Cost Query Pro

An infrastructure cost intelligence platform designed to transform historical construction cost data into actionable engineering insights through analytics, automation, and applied machine learning.

Built for engineers, estimators, project managers, and decision-makers, Cost Query Pro is designed to centralize infrastructure cost data, improve estimate consistency, and provide a foundation for predictive cost analytics.

Status: 🟢 Active Development

Primary Competencies: Infrastructure Analytics • Data Engineering • Backend Development • Cost Intelligence • API Design

• Project and bid item data management
• CSV and Excel import with schema validation
• Duplicate detection and data quality reporting
• RESTful APIs for querying infrastructure cost data
• Secure authentication and role-based access
• AI-assisted scope analysis and WBS generation
• Export pipelines for downstream reporting and analysis

• Historical cost normalization
• Regional cost modeling
• Bid item anomaly detection
• Predictive cost estimation
• Infrastructure cost forecasting
• Interactive dashboards and reporting
• Decision-support analytics

• Transforming historical bid data into actionable engineering intelligence
• Building scalable data pipelines for infrastructure analytics
• Developing maintainable backend architecture that supports analytical workflows
• Applying software engineering principles to improve engineering estimating processes

Goal: Develop a comprehensive infrastructure cost intelligence platform that enables engineers and organizations to analyze historical costs, improve estimating accuracy, and support data-driven project planning through modern analytics.

Long-Term Vision

Develop a production-ready infrastructure analytics platform that enables utilities to prioritize asset replacement, quantify infrastructure risk, and support engineering decision-making through applied machine learning and modern software engineering.

Lift Station Predictive Maintenance

An applied machine learning project focused on developing predictive maintenance strategies for municipal wastewater lift stations using SCADA data, operational metrics, and engineering knowledge.

The project explores how utilities can transition from reactive maintenance toward data-driven, risk-based asset management by identifying early indicators of equipment degradation and system performance changes.

Status: 🟡 Planning & Architecture

Primary Competencies: Infrastructure Analytics • Predictive Maintenance • Time-Series Analysis • Applied Machine Learning • SCADA Analytics

• System architecture and project planning
• Data strategy development
• Repository structure and engineering standards
• Machine learning workflow design
• Identification of key operational metrics and engineering features

• SCADA data ingestion and preprocessing
• Time-series feature engineering
• Pump performance monitoring
• Wet-well behavior analysis
• Failure prediction and risk scoring
• Operational dashboards and engineering visualizations
• Predictive maintenance recommendations

• Applying machine learning to improve infrastructure reliability
• Combining operational data with engineering expertise
• Developing practical analytical tools for municipal utilities
• Supporting proactive maintenance through data-driven decision-making

Goal: Develop a predictive maintenance platform that helps utilities improve reliability, reduce unplanned downtime, optimize maintenance planning, and extend infrastructure asset life through engineering analytics.

Together, these flagship projects represent different facets of the same long-term vision: applying engineering expertise, modern software development, and data-driven analytics to improve how infrastructure is designed, operated, and maintained.

While each project can stand on its own, they are intended to complement one another and collectively demonstrate the broader direction of my work in Infrastructure Analytics.

Not every engineering idea begins as a flagship project.

Some start as focused investigations into new analytical techniques, mathematical concepts, software design patterns, or emerging technologies before eventually influencing larger engineering platforms.

These projects provide an opportunity to explore new ideas, validate approaches, and expand my understanding of topics that may shape future infrastructure analytics applications.

While smaller in scope, each project represents an intentional investment in continuous learning and technical growth.

An applied machine learning project exploring credit risk prediction using supervised learning techniques and SHAP-based model explainability.

Although the application domain differs from infrastructure, the project provides valuable experience with classification models, feature engineering, model interpretation, and deployment concepts that are directly transferable to future infrastructure analytics projects.

Focus Areas

• Classification modeling
• Model explainability (SHAP)
• Feature engineering
• Applied machine learning

An exploratory data science project applying analytical techniques to publicly available environmental and climate datasets.

The project emphasizes data exploration, visualization, and analytical reasoning while providing experience working with complex real-world datasets.

Focus Areas

• Exploratory data analysis
• Environmental datasets
• Data visualization
• Scientific computing

A collection of small projects translating mathematical concepts into clean, well-structured Python implementations.

The objective is to strengthen mathematical intuition while exploring how numerical methods, probability, statistics, and engineering principles can be expressed through software.

Focus Areas

• Mathematical programming
• Numerical methods
• Probability & statistics
• Python fundamentals

These research projects often serve as testing grounds for ideas that may eventually become part of larger engineering applications. Exploring new techniques in smaller, focused environments allows me to evaluate concepts before integrating them into production-oriented projects.

Not every project has to solve a municipal infrastructure problem.

Some projects are simply opportunities to explore new technologies, experiment with interesting ideas, or satisfy a curiosity that's been sitting in the back of my mind. While these projects may be more lighthearted, they often introduce new techniques, architectures, or problem-solving approaches that eventually influence my professional work.

After all, some of the best learning happens when you're building something simply because it sounds interesting.

A probabilistic simulation exploring one of science's most fascinating questions:

If intelligent life should be common throughout the universe... why haven't we found any evidence of it?

Inspired by the Drake Equation and probabilistic modeling techniques, this project investigates how different assumptions influence the likelihood of technologically advanced civilizations emerging across the galaxy.

Although far removed from civil engineering, the project provides an engaging way to explore simulation modeling, probability, sensitivity analysis, and scientific computing.

Focus Areas

• Monte Carlo simulation
• Probabilistic modeling
• Sensitivity analysis
• Scientific computing

A playful full-stack web application built to explore modern web development using Next.js, edge databases, authentication, analytics, and premium user experiences.

Originally created as a fun project, it quickly became a practical environment for experimenting with modern application architecture and frontend development techniques that differ from my backend-focused engineering projects.

Focus Areas

• Next.js
• Edge databases
• Authentication
• Modern web architecture
• Full-stack development

While these projects may sit outside my primary engineering focus, they reflect something I value deeply: curiosity.

Whether I'm exploring hydraulic modeling, infrastructure analytics, probability, or even the possibility of intelligent life elsewhere in the universe, I enjoy learning how complex systems behave and how software can help us better understand them.

As my portfolio has grown, I've organized related projects into dedicated GitHub organizations based on their primary focus. This structure keeps repositories organized while reflecting the broader areas of work I'm actively developing.

Although each organization has a distinct purpose, they all contribute toward the same long-term vision of applying engineering, analytics, and software development to solve real-world infrastructure challenges.

Together, these organizations document my ongoing journey toward building practical engineering software and infrastructure analytics platforms that combine traditional engineering expertise with modern software development and applied machine learning.

Infrastructure engineering is entering an exciting period of transformation.

Advances in data availability, cloud computing, artificial intelligence, and modern software development are creating opportunities to improve how infrastructure is designed, operated, maintained, and renewed. I believe the future belongs to organizations that successfully combine engineering expertise with data-driven decision-making.

As I continue developing this portfolio, my goal is to contribute to that future by building practical engineering software that helps infrastructure professionals solve real-world problems.

I'm particularly interested in projects involving:

• 🏗️ Infrastructure Analytics
• 🧠 Applied Machine Learning
• 📊 Asset Risk Modeling
• ⚙️ Engineering Software Development
• 📡 Predictive Maintenance
• 📈 Decision-Support Systems
• 🌎 Digital Twins
• 💧 Water & Wastewater Infrastructure

While many of the projects in this portfolio are independent today, my long-term vision is to bring them together into a connected ecosystem of engineering tools that help infrastructure owners make better decisions through software, analytics, and engineering expertise.

The journey is just getting started, and I'm excited to continue learning, building, and contributing to the future of infrastructure engineering.

📬 brice@devbybrice.com | 🌐 https://www.devbybrice.com | 🔗 LinkedIn

Exploring the future of Infrastructure Analytics through engineering, software, and applied machine learning.