Hi, I'm Abasar Paudyal
Computer Engineering graduate with experience building scalable applications and taking machine learning models from research and training through to production deployment.
Experience
Associate Software Engineer
Ycotek
August 2026 - Present
AI/ML Engineer
Prixa Technologies
- Built computer vision pipelines for document forgery detection and face liveness/anti-spoofing detection.
- Fine-tuned ASR models and integrated TTS systems for Nepali-language conversational interfaces.
- Designed multi-agent workflows to coordinate specialized model components across the vision, speech, and language pipelines.
- Implemented RAG pipelines for improved retrieval and response accuracy for a Nepali-language assistant.
Jan 2026 - July 2026
Software Developer Intern
Danphe Innovations
- Developed responsive web applications using Next.js and Tailwind CSS with scalable frontend architecture.
- Integrated REST APIs to connect frontend components with backend services, ensuring reliable data flow across the application.
- Automated build and deployment using CI/CD pipelines and Docker for containerization.
- Optimized application performance and maintained code quality through modular architecture and testing practices.
Sept 2025 - Dec 2025
Education
Institute of Engineering(IOE), Thapathali Campus
Bachelors in Computer Engineering
April 2022 - April 2026
85.81 %
V.S. Niketan Secondary School
Higher Secondary in Science
June 2019 - Dec 2021
3.71 GPA
Skills
Programming Languages
Libraries and Frameworks
Databases
Tools
Projects
Monocular SLAM and NeRF Fusion for 3D Reconstruction
Real-time 3D reconstruction pipeline combining monocular SLAM, depth estimation and NeRF.
Nepali Multimodal Large Language Model
Multimodal system that generates conceptual Nepali text from text, image or combined inputs.
Intelligent Research Paper Analysis System
NLP system that identifies potential research gaps from scientific titles and abstracts.
Enhanced IOT Fire Prediction
Predictive IoT pipeline for fire detection using sensor data and machine learning models.
Care Connect
Hospital appointment system with role-based access, real-time doctor availability and PostgreSQL.
ChatSphere
Real-time messaging web app with live conversations, user management and story/status features.
Disaster Image Classification
Fine-tuned ResNet50 pipeline for classifying disaster and non-disaster Twitter images and tweets.
CLI Chat Application
CLI chat app with real-time messaging, custom B-tree database, secure authentication and CSV export.
BeFit
Dynamic workout planner with customizable exercise plans, responsive UI and client-side filtering.
Certifications & Publications
An Enhanced IoT Fire Detection Pipeline using Clustering-Based Resampling and Ensemble Learning
Digital Nepal Conclave (DNC), 2025
Supervised Machine Learning: Regression and Classification
DeepLearning.AI
Advanced Learning Algorithms
DeepLearning.AI
Unsupervised Learning, Recommenders, Reinforcement Learning
DeepLearning.AI
LOCUS Software Fellowship
LOCUS Pulchowk
Get In Touch
Have a project in mind or want to collaborate? Feel free to reach out!