Machine learning · Neuroimaging · Interpretable AI

Hello, I’m

Garvit Joshi

Doctoral Researcher in Systems Neuroscience

I build interpretable and reproducible models for understanding relationships between the human brain, behaviour, and personality.

Portrait of Garvit Joshi
Dresden, Germany · TU Dresden
Scroll down

Profile

Research with clarity and practical purpose

I am a doctoral researcher and research associate at Technische Universität Dresden. My work combines machine learning, statistical modelling, and neuroimaging to study relationships between brain connectivity, human behaviour, and personality.

I care about interpretable models, reproducible research, and scientific tools that help people explore complex results. My background spans computational logic, medical-image analysis, geospatial modelling, autonomous systems, and human-centred prototyping.

Research question

How can complex brain models become more interpretable?

Methods

Machine learning, statistics, fMRI, structural MRI

Principles

Interpretability, reproducibility, open science

Career

Research experience

Nov 2022–present

Technische Universität Dresden

Research Associate & Doctoral Researcher

Doctoral research in Systems Neuroscience since September 2023, embedded in TRR 265.

  • Model relationships between functional and structural MRI, human behaviour, and personality.
  • Develop interpretable machine-learning workflows with an emphasis on reproducibility and scientific software.
  • Neuroimaging
  • Machine Learning
  • Statistics
  • Open Science
Jun–Oct 2022

Fraunhofer IVI, Dresden

Research Assistant — Vehicle Control & Sensor Systems

Contributed to helyOS, an open-source platform for coordinating autonomous vehicles.

  • Adapted software for autonomous lawnmowers, route planning, and multi-vehicle operation.
  • Produced technical documentation and supported systems integration.
  • helyOS
  • Route Planning
  • Systems Integration
  • Open Source
Apr 2021–Mar 2022

Fraunhofer IVI, Dresden

Research Assistant — Geospatial Data Analysis

Analysed road-accident data and helped identify potentially hazardous road segments.

  • Developed classification and geospatial statistical methods for accident-risk analysis.
  • Built a Django web application with an interactive map for communicating results.
  • Django
  • Geospatial Analysis
  • Classification
  • Data Visualisation

Selected work

Research & projects

Research software · Neuroimaging

ConnExplorer

Contributed to an interactive web tool linking brain-connectivity matrices with anatomical regions and network-level visualisations.

  • Connectivity
  • Visualisation
  • Web Development
  • Open Science
Visit project

Earlier research · Technische Universität Dresden

SFB 940 & Project Z2 Core Imaging Facility

Earlier work supporting research on volition and cognitive control, including deep-learning analysis, scientific software, and neuroimaging infrastructure.

  • SFB 940
  • Neuroimaging
  • Deep Learning
  • Scientific Software
Visit project

Fraunhofer IVI

Autonomous Mobility with helyOS

Extended an open-source coordination platform for autonomous lawnmowers, route planning, and multi-vehicle operation.

  • MLOps
  • Route Planning
  • Systems Integration
  • Documentation
Visit project

Imagine Cup · Helios · Science communication

Human-centred Technology Prototypes

Built an Alzheimer-support application as a Microsoft Imagine Cup international finalist, worked on the Raspberry Pi-based Helios assistive device, and developed a VR brain-research exhibit shown at the German Hygiene Museum Dresden.

  • Computer Vision
  • Raspberry Pi
  • Virtual Reality
  • Prototyping

Transparent research

Research & Open Science

OSF preregistration

Improving Interpretability of Complex fMRI Models of Human Behavior

A preregistered part of the doctoral research programme, documenting hypotheses and analysis plans before confirmatory analysis.

View on OSF

Toolkit

Methods & technologies

Programming languages

A radial view of the programming languages in my technical toolkit.

Machine Learning & Data

  • Python
  • PyTorch
  • TensorFlow
  • Statistical Modelling
  • Deep Learning
  • Graph Neural Networks
  • BERT
  • BioBERT

Neuroimaging & Visualisation

  • fMRI
  • Structural MRI
  • FreeSurfer
  • SPM
  • Medical-image Segmentation
  • Connectivity Analysis
  • Virtual Reality

Software & Infrastructure

  • Git
  • CI/CD
  • Linux
  • CUDA/HPC
  • Django
  • Flask
  • Jupyter
  • SQL
  • Raspberry Pi

Logic & Knowledge

  • Computational Logic
  • Answer Set Programming
  • Prolog
  • OWL
  • SPARQL
  • Ontology Engineering

Languages

Hindi
Native
English
Professional working language
German
Goethe B1 certified; C1.1 academic course completed
Japanese & Czech
Basic knowledge

Academic background

Education

Sep 2018–Apr 2022

Technische Universität Dresden

MSc Computational Logic

Interdisciplinary programme spanning artificial intelligence, formal logic, mathematics, and computer science.

  • Master’s thesis at NCT Dresden on weakly supervised deep-learning methods for medical-image segmentation.
  • Coursework in computer vision, knowledge representation, ontology engineering, constraint programming, and complexity theory.

Jul 2014–Jul 2018

Chameli Devi Group of Institutions / RGPV

BE Computer Science and Engineering

Broad engineering foundation in software development, artificial intelligence, databases, networks, algorithms, and embedded systems.

  • Bachelor’s thesis on MIDI music generation using recurrent neural networks and LSTMs.

Get in touch

Let’s connect

I’m open to research collaborations and opportunities involving machine learning, neuroimaging, data science, and scientific software.