Divya Teripilli

AI/ML Engineer

Visakhapatnam $48/hr Hybrid Full-time 1+ yrs exp 92 · Outstanding

About

AI/ML Engineer with hands-on experience building LLM applications and RAG pipelines. Skilled in Python, TensorFlow, and data visualization, delivering practical solutions with measurable outcomes.

Skills & Expertise (25)

TensorFlow Intermediate

7.1/10

Years Exp

PyTorch Intermediate

7.1/10

Years Exp

scikit-learn Intermediate

7.1/10

Years Exp

Pandas Intermediate

7.1/10

Years Exp

NumPy Intermediate

7.1/10

Years Exp

FAISS Intermediate

7.1/10

Years Exp

ChromaDB Intermediate

7.1/10

Years Exp

Python Intermediate

7.1/10

Years Exp

Google Cloud Intermediate

6.4/10

Years Exp

Agile Intermediate

6.4/10

Years Exp

Kubernetes Intermediate

6.4/10

Years Exp

Azure Intermediate

6.4/10

Years Exp

AWS Intermediate

6.4/10

Years Exp

Unit Testing Intermediate

6.4/10

Years Exp

OOP Intermediate

6.4/10

Years Exp

REST APIs Intermediate

6.4/10

Years Exp

Docker Intermediate

6.4/10

Years Exp

MySql Intermediate

6.4/10

Years Exp

SQL Intermediate

6.4/10

Years Exp

CSS Intermediate

5.0/10

Years Exp

LINUX Intermediate

5.0/10

Years Exp

HTML Intermediate

5.0/10

Years Exp

Artificial Intelligence and Machine Learning LLMs / Generative AI (Exposure) RAG (Retrieval-Augmented Generation)

Work Experience

AI/ML Engineer

Growww Technologies Pvt. Ltd.

Aug 2025 - Present

Built and deployed an LLM-based chatbot using Google Gemini and LangChain with a full RAG pipeline — responses are grounded in retrieved documents stored as vector embeddings in FAISS. Implemented a query classifier that identifies the subject domain before retrieval, so the search is scoped to the right documents each time. Used ChromaDB alongside FAISS for persistent vector storage, testing retrieval performance across repeated sessions without re-embedding. Set up conversation memory so the bot maintains context across full session follow-up questions work naturally without the user having to repeat themselves. Wrote prompt templates to keep responses structured and consistent; added fallback handling for out-of-scope queries. Tested against 200 sample questions relevance rate above 90%, retrieval responses under 2 seconds. Deployed via Streamlit; accessible from a browser with no setup required on the user's end. Built ML models for internal use cases covering classification, regression, and clustering. Created Power BI dashboards used weekly by non-technical team members for performance tracking. Worked with the business team directly to understand requirements before starting any build. During my Data Science internship, I worked on an AI/ML chatbot project designed to help employees quickly access company information. I developed the chatbot using Python, Gemini, LangChain, and RAG. I created a pipeline where company documents were processed, converted into chunks, and transformed into embeddings. These embeddings were stored in a vector database such as FAISS or ChromaDB. When a user asks a question, the system retrieves the most relevant information and passes it to the Gemini LLM to generate a grounded response. I also implemented query classification, prompt templates, fallback handling, and Redis-based conversation memory. I tested the chatbot with around 200 questions and achieved more than 90% relevance, with retrieval responses generally under two seconds. I also used Streamlit for the user interface and FastAPI for backend APIs.

DATA SCIENCE INTENSHIP

EXCELR

MAR 2 - JULY 30

Worked on data cleaning, preprocessing, exploratory data analysis (EDA), and feature engineering using Python, Pandas, and NumPy. Built and evaluated Machine Learning models including Logistic Regression, Decision Tree, Random Forest, KNN, SVM, and XGBoost. Performed data visualization using Matplotlib and Seaborn to identify trends, patterns, and relationships in datasets. Applied K-Means clustering and PCA for customer segmentation and dimensionality reduction. Developed an LLM-based chatbot using Gemini, LangChain, RAG, and FAISS/ChromaDB to retrieve information from company documents. Built interactive Streamlit dashboards for presenting machine-learning predictions and business insights. Worked with SQL/MySQL for querying, filtering, joining, and analyzing structured data. Participated in model validation, performance evaluation, and improving model accuracy through feature engineering and hyperparameter tuning.