I’m a Software Engineer | Data Scientist | Data Engineer who enjoys building intelligent systems for messy, real-world problems — traffic prediction, incident response, conversational AI safety, and scalable data platforms.
Most of what I build comes back to one question:
How does a system know when to trust itself, and when not to?
🎓 B.Sc. Data Science — Bon Secours College for Women, Bharathidasan University CGPA: 9.3 / 10
🎓 Currently pursuing M.Sc. Data Science at SASTRA Deemed University
💻 Interested in software engineering, backend systems, data engineering, machine learning, Bayesian inference, and AI safety
📊 I enjoy solving ambiguous real-world problems where data, systems, and decision-making come together
🤝 Team Lead for the Meta PyTorch OpenEnv Challenge, working across reviewing, mentoring, debugging, and technical coordination
🏁 National-level hackathon , including Microsoft Imagine Cup, UIDAI Hackathon, and other innovation challenges
I build backend and system-oriented applications using APIs, databases, clean architecture, and reliable engineering practices.
I work on machine learning, Bayesian inference, prediction systems, uncertainty modeling, and decision-support applications.
I design data pipelines, ingestion workflows, validation layers, analytics-ready datasets, and scalable data systems.
I’m currently carrying the same “know what you don’t know” thinking from traffic prediction and incident response into conversational AI and reliable ML systems.
-
Dialogflow Sentinel A Dialogflow CX safety scanner paired with a Bayesian escalation-confidence webhook.
-
PTIS v2.0 A Bayesian traffic inference system focused on congestion prediction, uncertainty-aware routing, and publication readiness.
-
ADVE: Anchor-Delta Video Embedding A training-free approach for reducing redundant video encoding while preserving semantic meaning.
-
Cloud Security for ML Systems Learning IAM, DevSecOps, secure routing, observability, and the SRE side of AI systems.
- Software Engineering
- Backend Development
- Data Engineering
- Machine Learning
- Bayesian Inference
- Conversational AI Safety
- Large Language Models
- Cloud Security
- DevSecOps
- Reliability Engineering
- Research Engineering
Languages: Python, SQL, Java, JavaScript
Backend: FastAPI, REST APIs, Spring Boot basics, API design
Data Engineering: PostgreSQL, ETL pipelines, data validation, data modeling, Power BI
ML / AI: Scikit-learn, PyTorch, TensorFlow, NLP, LLM workflows
Cloud & DevOps: Git, GitHub Actions, Docker, Linux, AWS basics
Exploring: Dialogflow CX, IAM, DevSecOps, secure routing, SRE practices
Building backend services, APIs, and system workflows that are practical, reliable, and maintainable.
Creating pipelines that clean, validate, transform, and prepare data for analytics and machine learning.
Building uncertainty-aware traffic prediction systems that reason about congestion, confidence, and routing decisions.
Designing tools that help conversational agents detect unsafe, uncertain, or escalation-worthy situations.
Researching training-free methods to reduce redundant visual encoding while preserving semantic meaning.
🌐 Portfolio: asmitha2025.github.io
💼 LinkedIn: asmitha-m-1695b331b
🤗 Hugging Face: Asmitha-28
📧 Email: asmitha8825@gmail.com
Always building, always questioning, always learning how systems behave when the real world gets messy.