AI / ML

Sehastrajit Selvachandran

AI Scholar & ML Engineer

I build reliable AI assistants and machine-learning systems that balance technical rigor with clear communication for non-technical stakeholders.

Innovate, Improve, Repeat.

View Resume

sselvach@asu.edu

Phone

+1 (602) 554-6637

Location

Tempe, AZ · Remote-friendly

linkedin.com/in/sehastrajit-s

GitHub

github.com/Sehastrajit

Applied AI Programs

12+

From campus copilots to medical imaging research.

Datasets stewarded

4TB

Structured experiment logs with reproducible pipelines.

Multi-disciplinary teams

5 countries

Cross-timezone collaboration and stakeholder demos.

Experience

AI-native Projects

TITANS: LLM-Based Conversational Agent

Custom transformer assistant with long-context memory for multi-step reasoning.

Transformers • Long-context memory • Prompt engineering

Implemented memory modules to sustain dialogue context across extended sessions.
Iterated on instruction tuning and evaluation rubrics to improve helpfulness and alignment.

Diabetic Retinopathy Detection

Hybrid ResNet–ViT classifier for medical imaging analysis.

TensorFlow • PyTorch • CNN • Vision Transformer

Built preprocessing and augmentation pipelines across multiple retinal datasets.
Logged experiments for reproducibility and ethical decision-making.

Blue Zone Longevity Analysis

End-to-end ML pipeline with Streamlit interface for longevity insights.

scikit-learn • Streamlit • Pandas

Translated model outputs into narratives and visualizations for non-technical users.
Designed guided question flow that mirrors an assistant-style experience.

AI Ops Stack

LLM workflow design & guardrails

Experiment telemetry + observability

Responsible ML documentation

Human-centered interfaces for AI tooling

LUNA

Sehas's PA

Luna Studio

Ask anything about my portfolio

Responses are grounded in the resume, experience timeline, and publications you see on this page.

Response from Luna will appear here.

Skills

AI & LLMs

LLM assistant design • Prompt & guardrail authoring • RAG workflows • Response evaluation

ML & Data

TensorFlow • PyTorch • scikit-learn • Pandas • Experiment design • Feature engineering

Product & Delivery

Workflow documentation • Stakeholder comms • Remote collaboration • Requirements gathering

Tools

Python • Streamlit • Jupyter • Matplotlib • Git • Linux

Education

M.S. Computer Science

Arizona State University • Tempe, AZ

Expected May 2027

Relevant coursework: Machine Learning, Data Processing at Scale, Knowledge Representation, Data Visualization

B.Tech Computer Science (AI & ML)

Vellore Institute of Technology • Chennai, India

Jul 2025

Relevant coursework: Deep Learning, Image Processing, Natural Language Processing

Volunteer & Sustainability

Program Coordinator – CSR Medical Camp · URC Construction (P) Ltd

Chennai, India

Apr 2023

Coordinated medical camp logistics for 40+ participants, including registration and scheduling.
Documented outcomes to strengthen community engagement around preventive care.

Environmental Volunteer · URC Construction (P) Ltd

Chennai, India

Jun 2024

Led outreach for Environmental Awareness Rally focused on waste management and plastic-free zones.
Engaged community members on sustainable practices and event logistics.

Publications

Peer-reviewed research and citations.

View Google Scholar

Pipelined Structure in the Classification of Skin Lesions Based on AlexNet CNN and SVM

IEEE Access · Apr 2024

10.1109/ACCESS.2024.10496674

Comparative Performance of Deep Learning Architectures for Diabetic Retinopathy Detection

Int. J. Ad Hoc Ubiquitous Comput. · Sep 2023

10.1504/IJAHUC.2023.133449

Advances and Future Perspectives in Parkinson’s Disease Detection

Evolutionary Intelligence · Jul 2023

10.1007/s12065-023-00859-7

Statistical Features from VMD Levels for Sleep Apnoea Classification

Int. J. Bioinformatics Res. Appl. · Sep 2024

10.1504/IJBRA.2024.141381