I design and build end-to-end intelligent systems—from data pipelines and backend architecture to applied machine learning models and research-backed prototypes.
My work sits at the intersection of:
- Software engineering
- Applied machine learning
- Systems thinking
- Research-to-production workflows
I optimize for long-term capability, not short-term visibility.
In practice, I:
- Build production-grade backend systems and APIs
- Design ML pipelines from raw data → inference → application
- Apply statistical learning to real-world, noisy domains (especially healthcare)
- Translate research ideas into deployable, constrained systems
I routinely bridge roles that are often siloed:
System engineer × ML engineer × data scientist
Backend & Systems
- Node.js, Express
- MongoDB
- Python: Django, FastAPI
- API-first design, authentication, scalability considerations
Data & Machine Learning
- Python, NumPy, Pandas, SciPy
- Scikit-learn
- Regression (linear & non-linear), classification
- Feature engineering, validation, model evaluation
I work from first principles:
- Data before models
- Systems before frameworks
- Understanding before optimization
- Research before scaling
This keeps my work adaptable across domains and resilient to tooling churn.
A research-driven system focused on:
- Blood glucose prediction
- Non-linear statistical modeling
- Reliable inference under noisy, real-world data
Initial experimental validations show ~X–Y% improvement over baseline statistical predictors under controlled conditions. The system is architected to evolve from experimentation to clinically relevant deployment, not as a demo artifact.
I approach AI as a scientific discipline, not a feature set.
Current interests include:
- Predictive healthcare analytics
- Applied statistical learning
- Model reliability and interpretability
- Translating mathematical assumptions into operational systems
My long-term trajectory includes PhD-level research in AI / Computational Intelligence, with emphasis on original contribution and practical relevance.
In parallel with ML systems, I explore infrastructure-level problems, including feasibility work on improving digital connectivity in Nepal. This reflects the same systems mindset: constraints, incentives, reliability, and long-term impact.
- AI / ML engineering roles
- Research collaborations
- System design problems with real constraints
- Work that compounds over years, not quarters
I’m not optimizing for visibility. I’m optimizing for depth, leverage, and durability.
If you value:
- Systems thinking
- Research-backed engineering
- Quiet execution over loud claims