Skip to content
View AbhiChanGit's full-sized avatar

Highlights

  • Pro

Block or report AbhiChanGit

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
AbhichanGit/README.md

Hi there πŸ‘‹, I'm Abhishek Chandra

Computer Science major at UC, building intelligent systems at the intersection of production software engineering and computer vision research.

πŸ”§ Currently Working On

πŸ“± Android Automated Bug Reproduction Framework (REBL)

Working alongside a research team validating and expanding REBL, an end-to-end, LLM-powered framework that automatically reproduces Android crash and non-crash functional bugs using raw textual report descriptions.

  • Active Focus: Testing bug reproduction robustness across varying physical device matrices, OS configurations, and edge environment constraints.
  • Technical Footprint: Evaluating feedback-driven context prompts against UI Automator2 structures to optimize multi-action sequence executions and circumvent traditional token threshold limitations.

πŸ§ͺ Previous Initiatives & Research Projects

πŸ€– Local RAG Enterprise Engine | UC-Intelligence-Hub

  • Architected a fully localized Retrieval-Augmented Generation (RAG) platform using FastAPI, LangChain, and ChromaDB to parse and query university partnership records securely.
  • Embedded an asynchronous SharePoint REST pipeline wrapped in Microsoft Entra ID authentication and coupled it with FlashRank document rerank algorithms to suppress LLM hallucination indexes.

πŸƒ Biomechanical Kinematics & Signal Processing | GripForce-LoadForce | QTM-Realtime-Kinematics

  • Data Transformation: Developed Python data transformation utilities to handle biomechanical signal data extraction, structuring time-series multi-sensor load distributions for force resistance analysis.
  • Real-Time Streaming: Built real-time asynchronous streaming scripts using qtm-rt to intercept live 3D marker coordinate trajectories directly from Qualisys Track Manager, mapping continuous hip and knee sagittal angles on the fly using vector dot-product geometry.

πŸ«€ Deep Learning Cardiac Segmentation | Computer Vision

  • Built and optimized deep 3D multi-class UNET segment models utilizing the MONAI framework to map structural heart features, achieving an average 0.88 Dice coefficient score and removing manual slicing latency bottlenecks.
  • Configured pipeline preprocessing criteria and experimented with zero-shot segmentation leveraging Meta’s Segment Anything Model (SAM 2) to accelerate diagnostic testing loops.

πŸ“¬ Let's collaborate!

Popular repositories Loading

  1. test_actions_repo test_actions_repo Public

    A repository to test my GitHub actions

  2. DB_Project DB_Project Public

    TypeScript

  3. langchain langchain Public

    Forked from langchain-ai/langchain

    πŸ¦œπŸ”— Build context-aware reasoning applications

    Jupyter Notebook

  4. AbhichanGit AbhichanGit Public

  5. ml-cv-lab ml-cv-lab Public

    HTML

  6. ReBL ReBL Public

    Forked from datareviewtest/ReBL

    Python