RealTime Context Engine  

system_architecture RealTime_Context_Engine

Interview Assistant (aka Job Bandit) is an end-to-end orchestration and experience layer designed for the high-stakes environment of technical interviews. Engineered with a Data Engineering first mindset, the application manages the entire lifecycle of interview data—from real-time screen/audio ingestion and high-fidelity RAG (Retrieval-Augmented Generation) context enrichment to the delivery of optimized, low-latency streaming responses.

Orchestration: A seamless bridge between a versatile Electron desktop shell and a high-performance FastAPI backend, ensuring local compute efficiency and secure model interaction.

Advanced RAG Pipeline: Dynamically grounds AI responses in user-specific professional contexts by injecting resume data and real-time job requirements into the LLM context window via automated data-layer mapping.

Unified Data Persistence: Features a resilient session management system, automated export pipelines for post-interview synthesis, and searchable historical logs with MDX support for complex system design review.

Secure Licensing Layer: Includes an integrated RSA-backed licensing system that ensures professional-grade accessibility control and secure software deployment.

The "Convo Tab" Experience: A floating glassmorphism window provides live-streaming AI insights, comprehensive performance telemetry (TTFT, token usage), and cumulative cost tracking for total transparency.

Seamless Human-AI Interaction: Adopts a minimalist, dark-mode-first aesthetic with an intuitive "Snip-and-Solve" interaction model (Ctrl+K/S), enabling instantaneous screen-based problem solving without breaking the user's flow.

High-Fidelity Feedback Loops: Tracks live metrics such as Time-To-First-Token (TTFT) to ensure the interface feels alive and responsive, aligning with the precision required for complex engineering and product design interviews.

Python JavaScript HTML5 CSS SASS Electron Glassmorphism UI GPT-4+ GPT-5+ OpenAI API Hardware Licensing Markdown
 Download Demo Release  Windows Only
SmartScreen may prompt you — this app uses a local developer certificate. If Windows flags it, click More info → Run anyway to proceed.

Cross-Platform Music Synchronization  

yt_sp

The “Monthly Music Sync” project is designed with the goal of synchronizing a user's liked songs between YouTube Music and Spotify platforms using Python and relevant APIs. The ELT pipeline is built using Python, AWS S3, YouTube Data API, Spotify API, and integrates with third-party libraries such as spotipy and youtube_dl. The project retrieves music data from YouTube Music and Spotify using their respective APIs and processes this data by extracting metadata and searching for corresponding tracks on Spotify. The processed data, specifically YouTube video IDs, are stored in an AWS S3 bucket for tracking synchronized songs.
The pipeline is automated to run monthly via Apache Airflow, ensuring periodic synchronization of liked songs.
Future enhancements include further integration with Apache Airflow for automation, development of a React App for a user-friendly interface, and enhanced error handling mechanisms.

Spotipy youtube_dl Pandas Plotly Python AWS S3 YouTube API Spotify API Apache Airflow React

Spotify User's "Liked Songs" Data exploration & visualization  

With the aim of retrieving the user's music data via Python's REST API and staging it to the Postgres server, the ELT pipeline has been designed using a combination of Python, Postgres database, Dagster for automation, docker for containerization, and Power BI for visualization. The data is transformed in a number of ways using pandas to produce useful insights that can be used to build PowerBI dashboards and visualizations. With Dagster, the pipeline is automated to run every month. The project's ultimate objective is to discover the user's monthly music preferences by analyzing particular characteristics, such as the recently released "audio-features" by Spotify.

Pandas Plotly urllib pandasql request Dagster Docker Container Power BI

About me.

I am a Data Engineer with 6+ years of experience specializing in the architecture and maintenance of high-stakes data pipelines within the healthcare, real estate, and telecom sectors. My expertise lies in bridging the gap between legacy infrastructure—like the complex EDI and CMS workflows I managed at Blue Cross Blue Shield—and modern cloud-native solutions using Azure Databricks and LLM orchestration.

I am a hands-on builder who values operational stability as much as innovation. I've led offshore teams through high-pressure recovery efforts and managed over 30 production deployments in environments where data accuracy is a regulatory requirement. I approach engineering with a product-focused mindset, recently developing custom AI-driven applications that utilize low-latency streaming and RSA security to solve real-world efficiency gaps.

Here is my downloadable resume.




Tools and Technologies:

Elements

Text

This isboldand this isstrong. This isitalicand this isemphasized. This issuperscripttext and this issubscripttext. This isunderlinedand this is code:for (;;) { ... }. Finally,this is a link.

Heading Level 2

Heading Level 3

Heading Level 4

Heading Level 5
Heading Level 6

Blockquote

Fringilla nisl. Donec accumsan interdum nisi, quis tincidunt felis sagittis eget tempus euismod. Vestibulum ante ipsum primis in faucibus vestibulum. Blandit adipiscing eu felis iaculis volutpat ac adipiscing accumsan faucibus. Vestibulum ante ipsum primis in faucibus lorem ipsum dolor sit amet nullam adipiscing eu felis.

Preformatted

i = 0;
while (!deck.isInOrder()) {
    print 'Iteration ' + i;
    deck.shuffle();
    i++;
}

print 'It took ' + i + ' iterations to sort the deck.';

Lists

Unordered

  • Dolor pulvinar etiam.
  • Sagittis adipiscing.
  • Felis enim feugiat.

Alternate

  • Dolor pulvinar etiam.
  • Sagittis adipiscing.
  • Felis enim feugiat.

Ordered

  1. Dolor pulvinar etiam.
  2. Etiam vel felis viverra.
  3. Felis enim feugiat.
  4. Dolor pulvinar etiam.
  5. Etiam vel felis lorem.
  6. Felis enim et feugiat.

Actions

Table

Default

Name Description Price
Item One Ante turpis integer aliquet porttitor. 29.99
Item Two Vis ac commodo adipiscing arcu aliquet. 19.99
Item Three Morbi faucibus arcu accumsan lorem. 29.99
Item Four Vitae integer tempus condimentum. 19.99
Item Five Ante turpis integer aliquet porttitor. 29.99
100.00

Alternate

Name Description Price
Item One Ante turpis integer aliquet porttitor. 29.99
Item Two Vis ac commodo adipiscing arcu aliquet. 19.99
Item Three Morbi faucibus arcu accumsan lorem. 29.99
Item Four Vitae integer tempus condimentum. 19.99
Item Five Ante turpis integer aliquet porttitor. 29.99
100.00

Buttons

  • Disabled
  • Disabled

Form