avatar

Deepak Choudhary

Research Assistant
Indian Institute of Science (IISc)
deepakcl (at) iisc.ac.in


Selected Projects

  1. P4-Netronome Workflow Setup & Testing
    GitHub
    • Carried out the hardware, software, and network topology setup from scratch to compile, deploy, and test P4 programs on Netronome SmartNIC
    • Established a network topology of workstations equipped with Netronome SmartNICs to successfully test the In-Band Network Telemetry (INT) and custom-scheduling P4 programs.
    • Automated the compilation and testing P4 programs with python and bash scripts.
    • Performed extensive latency measurements with the topology to design deterministic scheduling algorithms.

  2. Data Visualization dashboards
    • Linked Plots Visualization using ParasolJS [code]: Interactive Web Application for Linked Parallel Coordinates Plots using an open source JavaScript visualization library called Parasol.
    • Data Visualization Web Application using PivotTable.js [code]: Minimalist UI where the user can upload a .csv dataset and start analyzing using the functionalities of pivot tables and charts.
    • Data About Data [code]: This web application gives insights about the dataset uploaded like numerical columns, categorical columns, empty cells, and a lot of other insightful data.

  3. Efficient RUG circuits designs using Quantum-dot Cellular Automata (QCA)
    Report
    • Worked on the Reversible Universal Gates (RUG) and designed combinational circuits using RUG.
    • Proposed novel approaches to design reversible gates, resulting in reductions in both area and power dissipation.
    • Presented a comprehensive evaluation of the proposed gates, demonstrating their superior performance compared to the traditional CMOS circuits.

  4. Language Translation Model for Coastal Kannada
    GitHub
    • Initiated and led the collection of a substantial Kannada dataset encompassing region-specific vocabulary, specifically focusing on the Coastal Kannada dialect.
    • Leveraged advanced web scraping techniques to efficiently source the region-specific vocabulary.
    • Acquired holistic skill-set regarding Machine Translation, building a dataset from scratch, and problem-solving abilities.