MS Data Science at WSU. Software Engineer at Ennovar. Previously at LIS Nepal - cut LLM inference time by 40%. Published researcher. Builder of things that last.
Computer Engineer turned Data Scientist. I build production ML systems that people actually use.
I graduated with a 3.88 CGPA from Kathmandu University and have since worked across production ML systems, mobile applications, and full-stack platforms in Nepal and the United States. My research in Nepal Lipi OCR was published at LREC-COLING 2024, which reflects how seriously I approach documented, rigorous work.
At LIS Nepal I shipped optimizations that reduced LLM inference time by 40% and cut API token costs by 35%. I now work at Ennovar, WSU's industry-facing software center, building backend services for university-industry platforms.
Outside engineering, I care about thoughtful design, clear communication, and building software that people actually want to use. I speak English, Nepali, and Newari natively.
Ennovar at Wichita State University
LIS Nepal Pvt. Ltd.
LIS Nepal Pvt. Ltd.
Wichita State University · Wichita, Kansas
Kathmandu University, School of Engineering · Dhulikhel, Kavre
Trinity International SS & College · Kathmandu
Meridian International School · Kathmandu
School Topper - 2073 B.S. Batch
Nepal Lipi Text Recognition
OCR system for complex Nepal Lipi script including conjuncts and modifiers. Curated a custom dataset of 72 characters across handwritten and typed text. Led to a published paper at LREC-COLING 2024.
Transfer Learning Pipeline
Full training pipeline with preprocessing, augmentation, batch normalization. Systematic experiments comparing CNNs vs. ResNet variants, analyzing overfitting and convergence.
Abstractive summarization via Transformer Encoder-Decoder with Masked Multi-Head Attention.
Cross-platform social app for cultural heritage revival. Real-time Firebase. Android & iOS.
Engineering economics tool with sensitivity analysis, break-even, financial ratios. iOS & Android.
Food delivery platform with cart, reservations, order tracking, and history-based recommendations.
Digitized ledger with multi-user access, audit trails, CSV export, and Caesar cipher encryption.
Core competencies built through production work, not just coursework.
Bold entries represent
primary expertise areas.
SIGUL-WS 2024 · Proceedings of the 3rd Annual Meeting of the Special Interest Group on Under-resourced Languages @ LREC-COLING 2024 · pp. 244–251 · May 2024 · Publisher: ELRA and ICCL
Certifications & Awards
Wichita State University · Issued May 2026 · Does not expire · Accredited by Higher Learning Commission
Verify Credential →Open to full-time engineering and data science roles, research collaborations, and interesting project conversations.