Sushit Pakrashy

AI Engineer building LLM agent systems and the infrastructure that keeps them reliable — published two open-source Python packages implementing peer-reviewed research, each backed by 200+ automated tests and CI/CD.

Experience spans multi-agent pipelines (LangGraph, RAG) and active open-source contribution to LiteLLM. Seeking a Summer AI Engineering internship.

AI/ML Engineer Intern at iNeuBytes (2026-06-29 – 2026-07-29)

• Building and evaluating CNN and LSTM models for image classification (CIFAR-10) and sentiment analysis, comparing custom architectures against classical baselines.
• Developing a full-stack ML/NLP web application with a Flask backend and public deployment.

Data Science & AI Intern at Thiranex (2026-05-23 – 2026-06-22)

• Built an XGBoost heart-disease risk model (~90% accuracy, 0.93 AUC-ROC) with SHAP explainability.
• Built a Random Forest content classifier on Netflix metadata (85.7% accuracy).

B.Tech in Electronics and Communication Engineering – Guru Nanak Institute of Technology, Kolkata (2027)