ANISH SHARMA
AI/ML Engineer, Educator & Researcher
Building intelligent systems that learn, adapt, and explain
// About Me
Who I Am
Computer Science educator and AI researcher specializing in Machine Learning engineering, diagnostic deep learning, and curriculum delivery. Experienced in designing end-to-end ML pipelines, detecting complex algorithmic bugs, and mentoring 150+ students in predictive modeling and data prep.
With a strong foundation in mathematics and computer science, I focus on creating ML pipelines that are not only accurate but also interpretable and production-ready. My work spans deep learning, explainable AI, and model validation integrity.
When I'm not training models, you'll find me contributing to open-source projects, writing technical blog posts, or exploring the latest research papers. Based in Kathmandu, Nepal.
10+
Projects
3+
Years Experience
15+
Technologies
// Skills
Tech Stack
ML & AI
Data Science
Software & Web
// Projects
Featured Work
EvalGuard
AI-powered Socratic ML debugging platform that identifies 'silent' ML bugs (data leakage, improper metrics on imbalanced sets, hyperparameter tuning on test sets) that produce misleading results without throwing runtime errors. Built static and dynamic analysis workflows to extract code-level evidence and auto-generate Socratic guidance for students.
NeuroLens
Diagnostic deep learning framework for early-stage dementia classification from brain MRI scans. Addressed severe class imbalance using Soft Voting Ensemble Learning and Weighted Cross-Entropy. Integrated Grad-CAM heatmaps for clinical interpretability. Achieved 93% overall accuracy and 100% recall on rare 'Moderate Demented' cases.
Anomaly Detection in Financial Services
Implemented an Anomaly Transformer on the BankSim dataset to detect fraudulent transactions in an unlabelled, highly skewed financial environment. Optimized validation thresholds via cross-validation, achieving an ROC AUC of 0.79 with high precision (98%) and recall (94%).
// Experience
Where I've Worked
Sep 2025 — Present
Graduate Teaching Assistant – Artificial Intelligence
Biratnagar International College (University of Wolverhampton)
Instruct 150+ undergraduate students on foundational and advanced data preprocessing, feature engineering, and model validation techniques using Python and Scikit-Learn. Design hands-on lab sessions focused on cleaning, transforming, and structuring raw datasets to ensure data integrity prior to model training. Mentor student research in predictive modeling, guiding them on rigorous evaluation metrics and preventing silent validation errors.
Aug 2025 — Nov 2025
Full Stack Developer
Flip IQ (Remote)
Managed end-to-end development modules for a property data domain system within a fast-paced remote engineering environment. Collaborated cross-functionally across time zones to scope technical requirements, maintain project timelines, and deliver reliable codebase updates. Executed proactive problem-solving strategies to resolve feature bottlenecks and optimize application workflow.
Education
BSc (Hons) in Computer Science
Herald College Kathmandu (University of Wolverhampton)
Pursued a Bachelor of Science (Hons) in Computer Science. Selected for competitive leadership programs including the Aspire Leaders Program and certified as a Career Coach (Foundation Level 1) by Mindler. Completed Python Specialization and SQL certifications.
// Volunteer & Community
Community Service
Hong Kong
Web Development Consultant & Developer
HKnutra
Designed and deployed a full-featured website integrated with a custom Content Management System (CMS) tailored for regional trail running organizations.
Kathmandu, Nepal
Web Developer
Nepal Netra Jyoti Sangh
Overhauled and modernized the central organizational website, integrating a streamlined CMS to improve content publishing efficiency for health services outreach.
// Contact
Let's Build Something Together
I'm always open to discussing new projects, creative ideas, or opportunities to be part of something amazing.
Download Resume