PhD, Bioscience Engineering2026 –
present
KU Leuven,
Leuven, Belgium
- Topic: Context-aware insect monitoring in agriculture using AI · KU Leuven Profile
- Promoter: Prof. Wouter Saeys
Research Interests
I am a PhD researcher at the MeBioS Division of KU Leuven, working on context-aware insect monitoring in agriculture using artificial intelligence. My broader interests include computer vision, deep learning, anomaly detection, adversarial robustness and cyber-physical security.
Education
M.Sc., Data Science2022 – 2025
Friedrich-Alexander-Universität
Erlangen-Nürnberg, Germany · Grade: 1.8 (German scale)
- Thesis: Enhancing Smart Grid Security: A Deep Learning Approach to Adversarial Intrusion Detection · Code · PDF
- Supervisors: Dr.-Ing. Loui Al Sardy, Mamdouh Muhammad, M.Sc.
- Focus: applied machine learning, deep learning, cyber-physical security
B.Sc., Computer Science and Information Technology2015 – 2019
Tribhuvan University,
Kathmandu, Nepal · 72.07/100 (First Division)
- Research project: Devanagari Handwritten Character Recognition
Publications
- R. Acharya, L. Al Sardy, M. Muhammad, R. German. ADVIS-G: An Adversarially Defended Intrusion Detection System for Smart Grids Using Deep Learning. KI – Künstliche Intelligenz, 2026.
- R. Acharya, L. Al Sardy, M. Muhammad. ROSAID-ST: Adversarially Robust Intrusion and Anomaly Detection in ROS Networks via Student-Teacher Deep Learning. International Journal of Advanced Robotic Systems, under review, 2026.
Research & Professional Experience
Master's Thesis Researcher and HiWiApr 2025
– Mar 2026
Chair for Computer
Science 7, FAU Erlangen-Nürnberg
- Developed ROSAID-ST and ADVIS-G, adversarially defended intrusion detection systems for smart-grid communication data.
- Prepared exercises and tutored students in the course Cybersecurity Fundamentals in Networking.
Working Student, Computer VisionMay 2023
– Mar 2026
Primetals Technologies
Germany GmbH, Erlangen
- Engineered industrial vision pipelines for detection, segmentation, tracking and anomaly analysis.
- Built synthetic data and annotation tools for OCR and segmentation in limited-data settings.
- Developed FastAPI/SQL evaluation workflows, saving 100+ hours of manual video analysis.
Associate Data ScientistAug 2021 – Aug
2022
Extensodata Pvt.
Ltd., Kathmandu
- Led a team of 3 building a business intelligence tool and fintech data pipelines for loan-risk analysis.
Freelancer2022; Mar 2026 – present
Upwork, remote
- Built a real-time alert-based strategy backtesting platform; delivered full-stack AI and computer-vision applications.
Additional experience
- diyo.ai, Unity3D Development Intern (Nov 2020 – Mar 2021)
- MPercept Technology, AI Developer Intern (Aug 2019 – Jan 2020)
Selected Open-Source Projects
- Session Feature Extractor (2026): PyPI tool for extracting image features from network traffic data.
- Image Baker (2025): annotation and synthetic data generation for anomaly detection.
- SmokeSim (2024): smoke simulation, image augmentation and segmentation.
- Neural Template Matching (2024): U-Net experiment for template matching on COCO.
Skills
- Programming
- Python, SQL, Bash, JavaScript (basic), C# (basic)
- Machine learning
- PyTorch, TensorFlow, Scikit-learn, HuggingFace, MMEngine, OpenCV
- ML systems
- Docker, Git, MLflow, DVC, ONNX, TensorRT, Airflow, FastAPI, Flask, Streamlit, AWS EC2, CI/CD
- Languages
- English (C1), German (A2), Nepali (native)
Beyond Research
Writing poems and stories, reading, running, mushroom hunting, and listening to (symphonic) metal music, among other things.