Jagarajupalli Manasa profile photo 1 on Tal

Jagarajupalli Manasa

Aspiring Data Analyst | Building skills in Power BI | SQL | Python | Advanced Excel & Azure |

Bengaluru

1+ year of work experience

Self-Employed
Sep 202511 mosPresent

Bangalore Rural, Karnataka, India

What I’m building An ML‑powered web app that recommends books using a hybrid approach—collaborative filtering plus content‑based similarity—to cut decision fatigue and improve discovery across genres. Core features (MVP) Onboarding: pick genres/authors + quick ratings to seed a profile. Home feed: For You, Because you liked…, Trending, Fresh picks. Search: title/author/description with genre/year/rating filters. Book page: cover, authors, description, tags, similar titles, rate/add to shelf. Feedback loop: ratings/shelves refine recommendations. Admin: CSV import (Goodbooks‑10k), one‑click retrain, metrics dashboard. Tech stack Data/ML: Python, pandas, scikit‑learn, TF‑IDF + cosine, item‑item CF (v1), path to SVD/KNN (v2). Backend: FastAPI (REST), background retraining, model registry. Frontend: React, responsive UI with carousels and instant search. Storage: PostgreSQL; versioned model/vectorizer artifacts. Ops: Health checks, env‑based config, basic observability. What’s done Requirements, schema, preprocessing plan. TF‑IDF pipeline and item‑item similarity design. Hybrid scoring (CF + content + popularity) with fallbacks. API contracts and UI wireframes (onboarding, home, book page). Admin flows (import, retrain, metrics) scoped. In progress Data cleaning and vectorizer/index build. Popularity aggregates, interaction thresholds. API stubs for recs and similar items. Evaluation scripts: RMSE/MAE, Precision@K/Recall@K/NDCG@K. Diversity reranking (MMR) and “Why this” explanations. Next up Full offline eval and metrics dashboard. A/B flags for hybrid weights and exploration. Optional phase‑2: Google Books covers, SendGrid weekly digests. Call for feedback Diversity metrics that correlate with reader satisfaction. Cold‑start UX patterns that work for book discovery. Open datasets with richer book metadata. #MachineLearning #RecommenderSystems #DataScience #AIProducts #Python #FastAPI #React #MVP #Books #EdTech
Frontend Developer (Personal Project)
May 20253 mosJul 2025

Bengaluru, Karnataka, India

Tech Stack: JavaScript, HTML, CSS, OpenAI APIs Tools Used: Bolt.new, Netlify Developed a productivity timer app based on the Pomodoro Technique to help users manage focused work sessions and breaks. Integrated AI-generated motivational prompts using OpenAI APIs to enhance user engagement. Built using Bolt.new’s no-code platform and deployed via Netlify with a responsive, user-friendly interface. Demonstrated skills in UI/UX, event-based JavaScript, and no-code deployment.
Jan 20256 mosJun 2025

Bengaluru, Karnataka, India

Tech Stack: React (simulated), Node.js (simulated), TypeScript, PostgreSQL, Tailwind CSS, JWT Tools Used: Bolt.new, PostgreSQL, Netlify Designed and deployed a no-code full-stack system for managing dairy inventory, supplier records, and order tracking. Simulated frontend (React) and backend (Node.js) architecture with secure JWT-based authentication. Incorporated AI-generated summaries for automated reporting using prompt engineering techniques. Published via Netlify, showcasing real-world application flow and database interaction without traditional coding.
UTL Technologies ltd.
Feb 20253 mosApr 2025

Bengaluru, Karnataka, India

* Delivered four ML modules in healthcare (diabetes, breast cancer, heart disease, drug-review sentiment) using Python, Scikit learn, NLTK. * Built reproducible pipelines: imputation for biologically implausible values, feature scaling, one hot encoding, TF IDF; stratified splitting. * Trained Logistic Regression, Random Forest, SVC, KNN, Multinomial NB; compared via accuracy, precision, recall, F1, and confusion matrices. * Produced interpretable visualizations (EDA, correlations, confusion matrices) and structured documentation for methods and outcomes. * Prioritized clinical relevance by focusing on recall for positive cases and transparent trade-offs in model selection.

Education

Bachelor of Computer Applications, Computer Science

Rjs first grade college

20223 yrs2025

Intermediate (BIPC)

Sri Chaitanya Junior College, Tanapalli, Tirupati

20212021

Join Jagarajupalli and 10k+ other professionals on

Tal Boss

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

Download Tal Boss

Jagarajupalli Manasa on Tal

Headline

Aspiring Data Analyst | Building skills in Power BI | SQL | Python | Advanced Excel & Azure |

About

Hello, I'm a BCA graduate with a passion for combining technology, creativity, and problem-solving, I thrive at the intersection of AI innovation and practical application. I’ve explored the power of AI/ML tools, prompt engineering, and no-code development, and applied these to real-world projects like a Pomodoro Timer with AI-generated motivation and an Efficient Dairy Management System simulating a full-stack architecture — all built and deployed using platforms like Bolt.new and Netlify. Through my internship at UTL Technologies, I worked with real datasets in healthcare, experimented with LLMs, and gained insight into how AI can transform everyday tasks. Whether it’s streamlining operations, building user-focused tools, or simplifying development with no-code solutions, I enjoy creating systems that are intelligent, efficient, and easy to use. *Prompt Engineering Guides:* Hands-on resources to help you master the art of crafting prompts for LLMs and generative AI tools (think: chatbots, creative writing, data analytics and beyond). - *AI & No-Code Toolkits:* Curated collections and step-by-step walkthroughs for building intelligent workflows, automating everyday tasks, and integrating AI into customer support, web development, and business operations—no advanced coding required. - *Personalized Learning Programs:* From live workshops to on-demand recorded sessions—learn how to deploy AI agents, automate data routines, and turn your wildest technical ideas into real projects. - *Consulting for Early-Stage Projects:* Got an idea but don’t know where to start? I offer 1:1 consulting for students and startups—guiding you on AI tool selection, building MVPs, and scaling project outcomes. 💡 Always learning. 🤝 Always building. 🚀 Always open to new opportunities.

Location

Bengaluru

Work experience

AI‑Driven Book Recommendation System (Ongoing Project) at Self-Employed, Sep 2025 - Present · 11 mos, Bangalore Rural, Karnataka, India, What I’m building An ML‑powered web app that recommends books using a hybrid approach—collaborative filtering plus content‑based similarity—to cut decision fatigue and improve discovery across genres. Core features (MVP) Onboarding: pick genres/authors + quick ratings to seed a profile. Home feed: For You, Because you liked…, Trending, Fresh picks. Search: title/author/description with genre/year/rating filters. Book page: cover, authors, description, tags, similar titles, rate/add to shelf. Feedback loop: ratings/shelves refine recommendations. Admin: CSV import (Goodbooks‑10k), one‑click retrain, metrics dashboard. Tech stack Data/ML: Python, pandas, scikit‑learn, TF‑IDF + cosine, item‑item CF (v1), path to SVD/KNN (v2). Backend: FastAPI (REST), background retraining, model registry. Frontend: React, responsive UI with carousels and instant search. Storage: PostgreSQL; versioned model/vectorizer artifacts. Ops: Health checks, env‑based config, basic observability. What’s done Requirements, schema, preprocessing plan. TF‑IDF pipeline and item‑item similarity design. Hybrid scoring (CF + content + popularity) with fallbacks. API contracts and UI wireframes (onboarding, home, book page). Admin flows (import, retrain, metrics) scoped. In progress Data cleaning and vectorizer/index build. Popularity aggregates, interaction thresholds. API stubs for recs and similar items. Evaluation scripts: RMSE/MAE, Precision@K/Recall@K/NDCG@K. Diversity reranking (MMR) and “Why this” explanations. Next up Full offline eval and metrics dashboard. A/B flags for hybrid weights and exploration. Optional phase‑2: Google Books covers, SendGrid weekly digests. Call for feedback Diversity metrics that correlate with reader satisfaction. Cold‑start UX patterns that work for book discovery. Open datasets with richer book metadata. #MachineLearning #RecommenderSystems #DataScience #AIProducts #Python #FastAPI #React #MVP #Books #EdTech

Pomodoro Timer with AI Integration at Frontend Developer (Personal Project), May 2025 - Jul 2025 · 3 mos, Bengaluru, Karnataka, India, Tech Stack: JavaScript, HTML, CSS, OpenAI APIs Tools Used: Bolt.new, Netlify Developed a productivity timer app based on the Pomodoro Technique to help users manage focused work sessions and breaks. Integrated AI-generated motivational prompts using OpenAI APIs to enhance user engagement. Built using Bolt.new’s no-code platform and deployed via Netlify with a responsive, user-friendly interface. Demonstrated skills in UI/UX, event-based JavaScript, and no-code deployment.

Efficient Dairy System (Academic Project) at RJS FIRST GRADE COLLEGE, Jan 2025 - Jun 2025 · 6 mos, Bengaluru, Karnataka, India, Tech Stack: React (simulated), Node.js (simulated), TypeScript, PostgreSQL, Tailwind CSS, JWT Tools Used: Bolt.new, PostgreSQL, Netlify Designed and deployed a no-code full-stack system for managing dairy inventory, supplier records, and order tracking. Simulated frontend (React) and backend (Node.js) architecture with secure JWT-based authentication. Incorporated AI-generated summaries for automated reporting using prompt engineering techniques. Published via Netlify, showcasing real-world application flow and database interaction without traditional coding.

Intern at UTL Technologies ltd., Feb 2025 - Apr 2025 · 3 mos, Bengaluru, Karnataka, India, * Delivered four ML modules in healthcare (diabetes, breast cancer, heart disease, drug-review sentiment) using Python, Scikit learn, NLTK. * Built reproducible pipelines: imputation for biologically implausible values, feature scaling, one hot encoding, TF IDF; stratified splitting. * Trained Logistic Regression, Random Forest, SVC, KNN, Multinomial NB; compared via accuracy, precision, recall, F1, and confusion matrices. * Produced interpretable visualizations (EDA, correlations, confusion matrices) and structured documentation for methods and outcomes. * Prioritized clinical relevance by focusing on recall for positive cases and transparent trade-offs in model selection.

Education

Bachelor of Computer Applications, Computer Science, Rjs first grade college

Intermediate (BIPC), Sri Chaitanya Junior College, Tanapalli, Tirupati

Jagarajupalli Manasa - AI‑Driven Book Recommendation System (Ongoing Project) in Bengaluru | Tal