Blog Posts

From Independent Components to a Unified System

This week marked a major milestone for L.A.A.R.K Lite as we successfully transitioned from isolated development modules to a fully connected system. What began as separate frontend, backend, database, and modeling efforts has now evolved into an integrated full-stack prototype.

System Integration: Connecting the Pieces

Our primary focus this week was establishing seamless communication between the frontend, backend, and database layers. Using Supabase as our cloud database solution, we connected our backend APIs to a live database and ensured that the frontend could dynamically fetch and display stored data.

This means that data entered through the user interface is now persistently stored, retrieved in real time, and reflected back in the dashboard — validating our end-to-end system architecture. Seeing live data flow through the application confirmed that our technical stack is functioning cohesively rather than as independent components.

Model Completion and Readiness for Deployment

Alongside system integration, we completed the development of our baseline machine learning model. Using the refined and merged dataset, we finalized feature selection, validated outputs, and ensured the model generates meaningful predictions aligned with our predictive maintenance objectives.

With the model completed and the infrastructure ready, we are now positioned to integrate predictive outputs directly into the dashboard experience.

Reflection: From Architecture to Implementation

This week highlighted the importance of modular design. Because each subsystem — frontend, backend, database, and ML — was developed with clear boundaries, integration was a matter of alignment rather than redesign.

The platform is no longer theoretical. It is operational.

Moving Forward

In the upcoming week, we will focus on:

  • Integrating model predictions into the live dashboard
  • Refining UI interactions for improved usability
  • Enhancing data validation and testing workflows
  • Preparing for MVP-level demonstration

With Supabase powering our database, APIs actively serving data, and our predictive model finalized, L.A.A.R.K Lite is steadily advancing toward a fully functional intelligent lighting asset management system.

Tagged as: , ,

Continuing Momentum at LuminaTech

This week at LuminaTech felt like a turning point.

What started as separate pieces, data, models, UI mockups, backend functions is slowly becoming a connected system. Our whiteboard says “LuminaTech,” but behind that name is real architectural progress.

From Idea to Baseline Model

We began the week by locking down something fundamental: clarity. We finalized our input features, confirmed our target definition, and built a baseline ensemble model to establish our first true performance benchmark. Instead of experimenting in multiple directions, we now have a clear reference point, a measurable starting line.

This baseline gives us something powerful:
a way to prove improvement, not just assume it.

Next step? Hyperparameter tuning. We’ll be running structured search strategies and comparing performance carefully, ensuring that any gain is meaningful and reproducible.

Strengthening the Backbone: Backend Progress

While modeling took shape, the backend architecture matured as well. We implemented and organized Supabase Edge Functions to better support workflow logic. Endpoints were structured more cleanly, making communication between the model layer and frontend more predictable and stable.

This is the invisible work that makes everything else possible, the difference between a demo and a system.

Bringing the Floor Plan to Life

On the frontend side, integration continued steadily.

We advanced our React interface and deepened the interaction between the floor plan and fixture-level views. Clicking on a fixture now feels less like navigating static UI and more like interacting with a living system. With updated code pushed to GitHub, collaboration is smoother and version control remains clean as multiple components evolve in parallel.

What’s Next: Moving Toward Full Integration

The upcoming week is focused on refinement and connection.

We’ll:

  • Tune the model and compare baseline vs optimized configurations
  • Stabilize backend responses for UI integration
  • Connect prediction outputs into the frontend flow
  • Begin validating the full pipeline

Data → Model → API → UI

Questions for Alignment

As we move toward deeper integration, we’re seeking clarification from our liaison engineer on two key areas:

  • What format should model outputs take for sponsor visibility?
    Risk score? Category? Time-to-failure?
  • Do we have sufficiently reliable maintenance/failure history to support stronger validation?

These answers will guide how we design both evaluation and user display.

Where We Stand

Hyperparameter tuning is defined.
UI integration is advancing.
Backend endpoints are taking shape.

LuminaTech is transitioning from isolated development into a cohesive predictive maintenance platform.

This week wasn’t just about code, it was about convergence.

Post-QRB1 Development and Data Integration

Following our Qualification Review Board (QRB 1) presentation last week, this week was focused on translating the panel’s feedback into concrete technical progress. The team shifted from presentation preparation to deeper implementation, system integration, and validation of our data and architecture decisions.

What We Accomplished This Week

This week, we completed the integration of weather data with the EnergyStar dataset, enabling us to better simulate real-world operational hours and environmental stress on lighting fixtures. This step directly supports the modeling assumptions discussed during QRB 1 and strengthens the realism of our approach.

We also connected our GitHub repository and Lovable-generated code to a local host environment, allowing the team to iterate more efficiently on UI development and frontend testing.

Technical Progress and Team Contributions

  • Nikhitha and Jugal worked on integrating weather data with the EnergyStar dataset to replicate operational conditions and environmental stress factors affecting light fixtures.
  • Satwik and Swetha focused on debugging technical issues in the Flask backend and successfully establishing a stable connection with the API.
  • Dhivya and Zhiyu advanced the React-based frontend architecture using Node and npm, while resolving JavaScript display-related errors.

As a team, we reviewed all provided PDFs and Excel sheets, extracting and organizing relevant information to support dataset preparation and future model development.

Reflection

This week highlighted the importance of aligning technical execution with clear system-level understanding. Building on QRB 1 feedback, we were more intentional about validating assumptions, structuring our data pipeline, and ensuring frontend, backend, and modeling components are progressing cohesively. The hands-on integration work helped surface issues early and improved cross-team communication.

Next Steps

In the coming week, our focus will be on:

  • Integrating the React frontend with the backend API.
  • Combining building information from PDFs and Excel sheets and extracting meaningful data for UI display.
  • Creating documentation for the prepared dataset and initiating model development based on this data.

Tagged as: , ,

Qualification Review Board (QRB1) Presentation and Feedback

This week our team delivered our Qualification Review Board (QRB 1) presentation, a key mid-semester milestone where we shared our current progress and approach with the panel and received their feedback.

What we presented

For QRB 1, we walked the panel through:

  • Our project goals and scope.
  • The current design and architecture.
  • Early progress on our prototype and implementation plan.
  • Key technical decisions and assumptions.

During the presentation, each team member explained their work areas and how they contribute to the overall IPPD project objectives. We focused on clearly communicating our design decisions, rationale, and next steps.

Feedback from the panel

The panel’s feedback was constructive and very informative. Major points included:

  • Strengths: They appreciated our problem framing, clarity of idea, and team coordination.
  • Areas to improve: Clarify parts of the technical workflow, expand on certain assumptions we made.
  • Technical guidance: Several suggestions focused on strengthening how we describe our data pipeline, modeling choices, and UI integration strategy.

This feedback helped us see some blind spots in our narrative and better prioritize what to refine before the next major review

Reflection

Receiving feedback directly from the panel was valuable not just for improving our presentation, but for strengthening our project’s direction and clarity of purpose. We realized that clear communication of technical choices matters as much as the technical work itself. This reflection has already influenced how we are shaping the next iteration of our design and prototype.

Next Steps

Going into the next week, our focus will be on:

  • Integrating the panel’s feedback into our project plan and prototype refinements
  • Increasing the cohesion of our presentation flow as a team
  • Advancing our technical implementation based on suggested improvements

The QRB 1 experience was an important checkpoint that helped us re-align our efforts and approach with expected expectations. We’re motivated by the feedback and excited to move forward!

Tagged as: ,

Building Momentum: Team 14 Moves from Planning to Real Product

As the new semester gets underway, Team 14 is excited to share a progress update that reflects how far we’ve come—and where we’re headed next. Building on the strong foundation established last semester, our focus has shifted from planning and system definition to bringing our platform to life.

From Structure to Execution

Following a successful SLDR and a well-defined project roadmap, this phase of the project has been about turning ideas into something tangible. The groundwork we laid earlier—clear goals, defined responsibilities, and a shared technical vision—has allowed us to move forward with confidence and direction.

Rather than starting from scratch, we are now executing against a clear plan, which has helped us maintain momentum and make meaningful progress early in the semester.

A Lovable Interface Is Taking Shape

One of our most exciting developments so far is the progress on our user interface. The team has been intentionally focused on creating an interface that is not only functional, but also intuitive, clean, and enjoyable to use.

The result is an early version of the platform that already feels cohesive and approachable—something we’re proud to continue refining. Prioritizing usability at this stage ensures that future features integrate smoothly and that the platform remains accessible to its intended users.

Early Access to API Data

Another key milestone is gaining initial access to relevant API data. This marks an important transition from working with assumptions and placeholders to engaging with real data sources. Having API connectivity at this stage allows us to better understand data structures, validate design decisions, and begin thinking more concretely about analytics, automation, and system behavior.

Tagged as: , ,

Kicking Off the Semester Together: Setting the Foundation for the Project

After a cheerful and refreshing winter break, our IPPD team kicked off the new semester this week with a renewed sense of energy and focus. Week 1 served as a fresh start for everyone, giving us the opportunity to reconnect as a team, realign on project goals, and set expectations for the work ahead. Since this was the first week back, our efforts were intentionally focused on planning and organization rather than technical development.

During this week, we worked collaboratively to establish a solid foundation for the semester. As a team, we prepared key planning components, including the project critical path, the work breakdown structure, and an initial assessment of risks and mitigation strategies. These activities helped us clarify responsibilities, understand task dependencies, and align on how our project will progress throughout the semester.

Although development has not started yet, this planning phase was essential in aligning our technical approach, business objectives, and project management goals. By defining roles, outlining dependencies, and identifying early risks, we established a clear direction for the semester and ensured that the team is prepared to transition into development in the coming weeks.

In addition to project planning, we attended a guest lecture by Bill McElroy from the Engineering Leadership Institute at the University of Florida. His talk focused on the standard of care and professional credibility in engineering. He emphasized that engineers are responsible not only for delivering technically correct solutions but also for making ethical, well-informed decisions that consider safety, reliability, and stakeholder trust. This perspective reinforced the importance of accountability and professionalism in our approach to the project.

Moving Forward

In the coming weeks, we will begin transitioning from planning to execution. Our focus will shift toward implementing the system architecture, developing the dashboard interface, and advancing the model components. The structure and roadmap established this week will help us move forward efficiently and reduce risks as the project becomes more technically involved. We also plan to carry forward the lessons from the guest lecture by ensuring that our design decisions, data handling practices, and documentation meet professional and ethical standards.

Reflection

Overall, Week 1 was a valuable and well-paced start to the semester. After the winter break, taking time to plan, align as a team, and reflect on professional responsibilities helped set a positive tone for the project. This week reinforced the idea that strong engineering outcomes are built on thoughtful planning, collaboration, and credibility. We are looking forward to building on this foundation as the semester progresses.

Shining the Light: Team 14 Wraps Up the Semester with SLDR Success

As the semester drew to a close, Team 14 brought months of collaboration, learning, and innovation together for our System Level Design Review (SLDR) presentation, a major milestone in the development of L.A.A.R.K: AI-Enabled Lighting Asset Management & Digital Twin Platform.

Bringing It All Together

Over the semester, our focus shifted from exploring possibilities to building structure, defining how our platform will manage lighting assets, predict maintenance needs, and support energy-efficient decisions. The SLDR gave us the space to reflect on this progress and demonstrate how our design is ready to move into development.

During the presentation, we walked our sponsor and faculty through our goals, progress, and lessons learned. The feedback we received highlighted our strong organization, clear communication, and forward-thinking approach to sustainability and smart-building innovation.

Lessons and Reflections

This semester was about more than just creating a product, it was about learning how to think like a team of engineers. From weekly meetings to planning sessions, we learned how to translate complex challenges into actionable steps and how to keep momentum even when data or direction was still evolving.

The SLDR was a proud moment that reminded us how much collaboration, adaptability, and perseverance can achieve. We closed the semester feeling confident in our vision and ready to bring our ideas to life in the spring.

Looking Ahead

Next semester, Team 14 will shift gears into implementation and testing, turning our design into a working prototype that facility managers can interact with and evaluate. We’re excited to take everything we’ve learned and continue building toward a system that makes lighting management smarter, simpler, and more sustainable.

L.A.A.R.K: Transforming Facility Lighting Management with AI-Enabled IntelligenceBuilding the Future of Smart Lighting: Inside LuminaTech’s

Team 14, LuminaTech, recently completed our SLDR presentation for the UF IPPD program, marking a major milestone in the development of L.A.A.R.K — an AI-Enabled Lighting Asset Management Platform designed to modernize facility operations.

Our team showcased how L.A.A.R.K integrates interactive floor-plan visualization, predictive maintenance powered by machine learning, and unified asset tracking to help organizations move from reactive repairs to proactive, data-driven decision-making. This combination of smart analytics and intuitive UI design aims to reduce maintenance costs, improve reliability, and simplify large-scale lighting management.

The feedback we received from faculty, coaches, and staff emphasized both the strength of our technical foundation and key areas to refine. Reviewers highlighted our clean UI prototype, clear problem framing, and strong understanding of predictive modeling. At the same time, they encouraged us to clarify our end-to-end data pipeline, strengthen our competitive differentiation, and improve speaker transitions for future presentations. We’ve already developed a concrete action plan—from adding a more detailed data flow diagram to rehearsing as a team and preparing fallback datasets to mitigate sponsor-data delays.

With this input, LuminaTech is entering Spring 2026 with renewed direction and momentum. Over the next semester, we will integrate real lighting-system API data, finalize our predictive model with automatic recalibration, enhance our web interface, and perform full system-level testing. We’re proud of the progress our team has made so far, and we’re excited to continue building a platform that brings meaningful, scalable value to facilities through innovation, intelligence, and design.

Turning PID Feedback into SLDR Momentum – Blog#11 11/14/2025

This week was about momentum and clarity. We aligned on the big picture, turning feedback into a sharper plan, and committed to a supervised learning direction for replacement prediction, with a stronger grounding in industry practices. That focus now guides our feature choices, validation approach, and prep for SLDR.

We sharpened our model direction this week: after PID feedback, we explored “mean lifetime” and industry practices (LM-80/TM-21) and chose a supervised approach for predicting replacement timing, using warranty and reference life as priors. We also clarified how vendors estimate expected life (LM-80 → TM-21), which guides our features and validation. On the UI side, Our team refined the flow (moving login confirmation), added colored bulb status to floor plans, linked product bands to purchasing, and surfaced brand/type with warranty life. Meanwhile, we compared backend frameworks to integrate models, database, and dashboard cleanly, sketching service boundaries and communication.

We finalized the SLDR structure and began drafting. The narrative follows Problem → Approach → Value, with concise sections on user needs and data sources (tested vs. reported specs). We’re assembling demo visuals: a clean system diagram (models ↔ database ↔ dashboard) and a floor-plan view where fixtures display predicted lifetime.

Moving Forward!!

we’re rolling updates into the Streamlit prototype based on recent feedback, for a smoother, and interactive UI. In parallel, we’ll continue model development with the new insights we gathered and visualize the model’s internal behavior and feature effects using Python-based explainability and plotting tools.

Reflection

This week we learned the power of focus and storytelling. Listening to feedback helped us see the big picture and choose a clearer path. We also learned how important it is to explain ideas simply, what matters to building managers, not just to us, and to keep the experience intuitive on the page.

Prototype Inspection Day (PID) – Blog#10 11/07/25

During​‍​‌‍​‍‌​‍​‌‍​‍‌ Prototype Inspection Day (PID), the LuminaTech team presented the prototype version of our predictive lighting system. We highlighted the on-demand UI dashboard, featuring role-based access, a floorplan viewer that is always up-to-date, and visual indicators for fixture lifetime and energy performance. The dashboard is a convenient tool for facility managers, as it allows them to keep a close eye on the lighting systems and, at the same time, recognize the occurrence of system failure in ​‍​‌‍​‍‌​‍​‌‍​‍‌advance.

We also presented the insights of our machine learning models, including Random Forest and XGBoost; these and EDA helped us identify the most important features in the Energy Star Dataset. This demonstrates our commitment to a data-driven approach while predicting the lifetime of a light fixture.

Feedback​‍​‌‍​‍‌​‍​‌‍​‍‌ from the liaison and coaches in terms of upgrading the prototype, user engagement, and system integration was of great value to us. Consequently, the team will make changes to the dashboard user interface, upgrade the alert and warranty functionalities, and begin working on the backend API layer to allow real-time data exchange between the ML models and the ​‍​‌‍​‍‌​‍​‌‍​‍‌dashboard.

MOVING FORWARD!!!

Following the feedback from PID, our team is focused on enhancing both the user experience and the technical foundation of our system.

  • Refining the UI with clearer visual indicators and warranty data.
  • Integrating real-time APIs for model communication.
  • Expanding our ML model to include energy optimization.

With these updates, LuminaTech is moving closer to a functional MVP that combines intelligent prediction with practical facility management.

Tagged as: , , , ,