About Experience Education Skills Projects Certifications Publications Connect

Bishwash Khanal

Jyväskylä, Finland

MSc AI Student & Research Assistant passionate about LLM fine-tuning, computer vision, and developing innovative AI solutions. Experience in model compression, 3D mesh processing, and cutting-edge deep learning applications.

Experience

My professional journey in AI & Computer Vision

Jan 2026 - Present

Research Assistant

University of Jyväskylä
Jyväskylä, Finland

Working on the AI Native Software Engineering (ANSE) Project, exploring AI-driven approaches to software engineering standards.

Learn more

Key Focus Areas

  • Researching AI-native software engineering processes
  • Exploring AI-generated code integration challenges
  • Collaborating with cross-functional teams
Mar 2024 - Aug 2025

Artificial Intelligence Engineer

OptiML Org (Part-time, Remote)
California, USA

Founding member working on LLM fine-tuning and compression for efficient large model deployment.

Learn more

Key Contributions

  • Developed custom pipeline for LLM fine-tuning & compression
  • Evaluated domain-specific model performance
  • Researched SOTA compression techniques

Technologies

PyTorch, Transformers, PEFT, bitsandbytes

Models

Llama, Qwen, Mistral, Gemma, Phi, Deepseek

Mar 2021 - Aug 2024

Computer Vision Engineer

E.K. Solutions Pvt. Ltd.
Lalitpur, Nepal

Led multiple computer vision projects including model compression, virtual tours, and facial recognition systems.

Learn more

Key Projects

  • Model Compression (2024): TensorRT quantization for Jetson AGX Orin
  • LLM Tools (2023-2024): RAG-based chatbot and CV analysis tools
  • Virtual Tours (2022-2024): 3D mesh generation from iPad Pro Lidar
  • Landmark Detection (2021): YOLO models for construction sites
  • Facial Check-In (2021): FaceNet-based attendance system
Jun 2023 - Oct 2023

Competitive Coder & Data Analyst

Scale AI (Freelancing)
California, USA

Freelance work through Upwork involving data annotations to train Generative AI models.

Learn more

Key Activities

  • Solving competitive programming problems for AI training
  • Evaluating and ranking AI model responses
  • Assessing factual accuracy of AI-generated text
During Studies

Teaching Assistant

Jacobs University Bremen
Bremen, Germany

Supported students in electronics and communications labs, graded assignments, and conducted tutorials.

Learn more

Courses

  • Intro to Electronics: KiCAD, LTSpice, PCB simulations
  • Communications Lab: C and MATLAB for DSP/communications
During Studies

Student Supervisor

IRC Library
Bremen, Germany

Supported students and patrons at the university library, trained new employees.

Education

Academic background

08.2024 - Present

M.Sc. in Artificial Intelligence

University of Jyväskylä
Jyväskylä, Finland

Advanced studies in AI with focus on deep learning, computer vision, NLP, and agent technology.

View details

Relevant Courses

  • Deep Learning for Cognitive Computing
  • Computer Vision and Image Analysis
  • Natural Language Processing
  • Collective Intelligence and Agent Technology
  • Semantic Web and Linked Data
09.2017 - 08.2020

B.Sc. Electrical & Computer Engineering

Jacobs University Bremen
Bremen, Germany

Graduated with Grade 1.73, Minor in Intelligent Mobile Systems.

View details

Relevant Courses

  • Signal Processing, Wireless Communication
  • Machine Learning, Computer Vision
  • Artificial Intelligence, Robotics
  • Information Theory, Embedded Systems

Thesis

Non-Linearity in Wireless Communications and Deep Learning

Skills & Technologies

Tools and technologies I work with

Programming

Python C C++ MATLAB Java

ML Frameworks

PyTorch TensorFlow Keras Transformers

ML Models

LLMs CNNs GANs YOLO Vision Transformers

Tools & Cloud

Docker FastAPI Git Linux

Data & CV

OpenCV Pandas NumPy MySQL

Environments

Linux Unix VS Code LaTeX

Featured Projects

Some of my recent work and contributions

Research

Nepali OCR using ViT

Vision transformers for Nepali text detection with RoBERTa-based tokenizer.

Python Transformers RoBERTa
Deep Learning

Face Generator (GANs)

Explored VAE, DCGAN, and WGAN for generating synthetic facial images.

PyTorch GANs VAE
Kaggle

ASL Recognition

TFLite model for ASL gestures. Ranked 371/1315 teams.

TensorFlow MediaPipe
Data Science

Bremen Big Data 2020

U-Net and LSTM for hand movement prediction from EMG data. Ranked 12/31.

TensorFlow U-Net LSTM
Thesis

Wireless & Deep Learning

Deep learning for modeling non-linearities in wireless communications.

Python Deep Learning
Research Paper

LLM Compression Impact

Evaluating compression techniques on LLM task-specific performance.

LLMs Quantization

Certifications

Specializations and courses completed

🧠
Deep Learning Specialization
Coursera
📝
NLP Specialization
Coursera
🎨
GAN Specialization
Coursera
⚙️
ML Specialization
Coursera
🚀
MLOps Specialization
Coursera
🔧
TensorFlow Developer
DeepLearning.AI
🗄️
Database Engineer
Meta

Publications

Research papers and academic contributions

2024
Alignment of 360° images with posed color images for locally accurate texturing
B. Khanal, M. Om, S. Rijal, V. Ojha
Frontiers in Computer Science
2024
Structure-preserving Planar Simplification for Indoor Environments
B. Khanal, S. Rijal, M. Awale, V. Ojha
arXiv
2024
Evaluating Compression Techniques on Task-Specific LLM Performance
B. Khanal, J. M. Capone
arXiv

Let's Connect

Find me across the web

I'm always interested in hearing about new projects, research opportunities, or just connecting with fellow AI enthusiasts!