Real Systems.
Built and Deployed.

Capture,process,and monitor high-speed live data reliably while maintaining accurate logs for diagnostics,validation,and later analysis.

Integrated an FPGA SoM for real-time data acquisition and processing with a software application that provided live monitoring,data visualization,and structured logging capabilities.

Delivered a unified platform for real-time data processing,continuous monitoring,and reliable logging,enabling faster diagnostics and more effective analysis of live system data.

Process audio with low latency while providing users with an easy way to tune sound characteristics without modifying the FPGA design.

Implemented the audio processing pipeline on an FPGA SoC to receive input sound files,apply configurable processing algorithms,and transmit the resulting audio to a speaker. Developed a software application to configure and adjust audio parameters in real time.

Delivered a configurable audio processing platform that enabled real-time sound processing and speaker playback,with software-based control over audio parameters for flexible tuning and testing.

Client onboarding required the research team to manually review long-form documents,identify relevant learning messages,interpret how each message aligned with DQ Lab’s framework,and prepare outputs for internal validation. The process was slow,repetitive,and difficult to scale because every new client document required manual extraction,interpretation,and alignment checks before program creators could move forward.

Designed and deployed a FastAPI-based document intelligence pipeline that used OCR to extract text from unstructured client documents,applied rule-based processing to clean and segment learning messages,and used BERT embeddings with semantic similarity to map extracted content against DQ Lab’s framework. The workflow stored structured outputs in MongoDB,supported cached remapping when researchers edited content,and kept a human-in-the-loop review layer so the system accelerated research without removing expert validation.

Reduced manual research effort by around 24 hours per week and turned a heavily manual onboarding process into a faster,repeatable workflow. The system improved the consistency of document interpretation,gave researchers structured outputs to review instead of starting from scratch,and created a scalable foundation for checking client-program alignment against the framework before the next stage of delivery.

The existing scoring workflow was not fast enough for a platform where schools,program creators,and internal teams needed timely assessment results. Score calculation,competency-level breakdowns,certificate generation,and dashboard reporting depended on reliable scoring logic and clean result data. As platform usage increased,the system needed to support client-specific scoring rules without slowing down the user and reporting experience.

Redesigned the scoring workflow using Python,Redis,and MongoDB to separate scoring rules,cached lookups,user results,and dashboard-ready outputs. The system supported competency-level calculations,client-specific scoring logic,certificate generation,and structured result data that could be consumed by product workflows and analytics dashboards. The focus was not only to make the API faster,but to make the scoring layer reliable enough to support reporting,certification,and downstream decision-making.

Reduced scoring API response time from 3–5 seconds to under 3 ms,making the assessment workflow significantly faster and more reliable. The rebuilt system supported real-time scoring,certificate generation,client-specific scoring rules,and dashboard-ready reporting,giving product and research teams a stronger foundation for assessment delivery and performance tracking.

Build a modern,responsive website that delivers a clear user experience while supporting reliable backend functionality,data handling,and seamless interaction between the interface and server.

Designed the frontend layout and user interface for desktop and mobile devices,then developed backend logic to handle form submissions,data processing,database operations,and communication between the user interface and application services.

Delivered a responsive and user-friendly website with a structured frontend design and scalable backend logic,enabling smooth user interactions,efficient data management,and reliable application performance.