Nishita Rana profile photo 1 on Tal

Nishita Rana

AI Engineer | Software Development Engineer | Generative AI, LLMs, RAG & Agentic AI Systems | Python, C++| NIT Andhra Pradesh ’26

Moradabad

< 1 year of work experience

IISER Bhopal
May 20253 mosJul 2025

Bhopal, Madhya Pradesh, India

• Developed a Human Action Recognition model using CNN for spatial feature extraction and LSTM for temporal sequence modeling. • Preprocessed the UCF50 dataset (frame extraction, normalization, segmentation, one-hot encoding) for efficient training. • Trained and evaluated the model to classify actions such as walking, jogging, sitting, and climbing stairs.

Education

National Institute of Technology, Andhra Pradesh

Bachelor of Technology - BTech, Electronics and communication engineering

National Institute of Technology, Andhra Pradesh

Intermediate, PCMB

Sacred Heart Convent School

Join Nishita and 10k+ other professionals on

Tal Boss

Tal is the app where bosses meet engineers & builders directly, without middlemen

Download Tal Boss

Nishita Rana on Tal

Headline

AI Engineer | Software Development Engineer | Generative AI, LLMs, RAG & Agentic AI Systems | Python, C++| NIT Andhra Pradesh ’26

About

I'm a final-year Electronics & Communication Engineering student at NIT Andhra Pradesh, passionate about building intelligent software systems at the intersection of AI and software engineering. My focus is on Generative AI, Retrieval-Augmented Generation (RAG), LLM applications, and scalable backend systems. I enjoy taking ideas from concept to deployment and solving real-world problems through technology. Recently, I built DocuMind AI, a production-grade RAG assistant using LangChain, ChromaDB, Ollama (Llama 3.2), and Streamlit. The system features incremental document ingestion using SHA-256 hashing, configurable retrieval strategies (Similarity Search, MMR, and Score Threshold), query rewriting, context compression, and source attribution for more accurate and transparent AI responses. As a Research Intern at IISER Bhopal, I developed an end-to-end Human Action Recognition system using ConvLSTM2D, handling the complete pipeline from video preprocessing and model design to training and evaluation, achieving ~77% test accuracy. My technical interests include: • Generative AI, LLMs, RAG & Agentic AI • Python, C++, SQL • Data Structures & Algorithms • OOP, DBMS, Operating Systems & Computer Networks • Deep Learning, NLP & Semantic Search • LangChain, ChromaDB, TensorFlow, Keras, Scikit-learn I'm currently seeking opportunities as an AI Engineer or Software Development Engineer where I can contribute to building scalable, intelligent, and user-focused products.

Location

Moradabad

Work experience

Machine Learning Intern at IISER Bhopal, May 2025 - Jul 2025 · 3 mos, Bhopal, Madhya Pradesh, India, • Developed a Human Action Recognition model using CNN for spatial feature extraction and LSTM for temporal sequence modeling. • Preprocessed the UCF50 dataset (frame extraction, normalization, segmentation, one-hot encoding) for efficient training. • Trained and evaluated the model to classify actions such as walking, jogging, sitting, and climbing stairs.

Education

Bachelor of Technology - BTech, Electronics and communication engineering, National Institute of Technology, Andhra Pradesh

Intermediate, PCMB, Sacred Heart Convent School

Nishita Rana - Machine Learning Intern in Moradabad | Tal