Sravani Surasani — AI/ML Engineer

// 01 — expertise

Skills & Tech Stack

🤖

AI / ML / Deep Learning

Machine Learning Deep Learning NLP Generative AI LLMs RAG Statistics

🧠

AI Systems & Architecture

Multi-Agent Systems MCP Vector DB Embeddings Semantic Search Prompt Engineering

🐍

Programming Languages

Python SQL R Java

📚

Libraries & Frameworks

Scikit-learn TensorFlow Pandas NumPy Streamlit ETL

🗄️

Databases & Storage

MySQL MongoDB SQLite Big Data

📊

Tools & Visualization

Power BI Excel Git Linux Data Visualization Feature Engineering Model Evaluation

// 02 — work

Featured Projects

PROJECT / 01

Multi-Agent RAG System with MCP

A production-ready, scalable multi-agent Retrieval-Augmented Generation pipeline for intelligent query answering, built with Model Context Protocol for seamless agent orchestration.

Architected a distributed multi-agent pipeline where specialized agents collaborate using MCP-based communication to handle complex queries end-to-end

Integrated vector database with dense embeddings and semantic search to retrieve contextually relevant information with high precision

Implemented intelligent query routing and agent orchestration, reducing hallucinations and improving answer accuracy significantly

Designed modular, production-grade architecture enabling seamless agent scaling and addition of new knowledge sources

Python LLMs RAG MCP Vector DB Embeddings Semantic Search

View on GitHub →

PROJECT / 02

AI Resume Screening System

An intelligent NLP-powered resume parsing and job-matching system that automates candidate ranking using embeddings and cosine similarity for precise role-fit evaluation.

Built end-to-end NLP pipeline for resume parsing — extracting skills, education, and experience entities with high recall

Applied sentence embeddings and cosine similarity scoring to objectively rank candidates against job descriptions

Automated the screening workflow, reducing manual review time and eliminating subjective bias in shortlisting

Designed an extensible system supporting multiple job roles and dynamic JD uploads for real-world HR teams

Python NLP Embeddings Cosine Similarity Scikit-learn

View on GitHub →

PROJECT / 03

Social Media Sentiment Analysis

A comprehensive sentiment classification system combining traditional ML and deep learning approaches to analyze public opinion from social media data at scale.

Designed and compared multiple models — from Logistic Regression to LSTM — benchmarking accuracy, F1-score, and inference speed

Built robust preprocessing pipeline: tokenization, stopword removal, lemmatization, and TF-IDF / word embedding feature extraction

Performed deep exploratory data analysis with visualizations to uncover sentiment trends across topics and timeframes

Evaluated models using confusion matrices, ROC curves, and precision-recall metrics for reliable production readiness

Python Deep Learning NLP TensorFlow Scikit-learn Pandas

View on GitHub →

// 03 — experience

Work Experience

AI/ML with Generative AI Trainer

Pumotechnovation

Feb 2026 — Present

Delivered comprehensive training programs on ML, NLP, Generative AI, and LLM applications to aspiring AI engineers
Mentored students in designing and building end-to-end RAG systems and autonomous AI agents using Python and vector databases
Designed hands-on, project-based curriculum using embeddings, vector databases, and real-world AI pipelines to maximize learning outcomes
Guided learners through industry-standard model evaluation, deployment practices, and responsible AI concepts
Bridged theory and practice — enabling students to build production-grade AI solutions from scratch

Sravani
Surasani

Skills & Tech Stack

Featured Projects

Work Experience

Academic Background

Certifications & Achievements

Let's Connect