Data Analyst · Business Analyst · AI enthusiast
Mumbai, India · B.Sc Computer Science, SIES College · Open to opportunities
I'm a data analyst and AI enthusiast from Mumbai with hands-on industry experience across the full data stack — from SQL querying and ETL pipelines to Power BI dashboards and NLP-driven analysis. Recently I've been building with LLMs, including a live AI sales intelligence app powered by Llama 3.3 (70B).
- 🔭 Currently open to Data Analyst / AI roles
- 💡 I love turning messy data into decisions stakeholders actually use
- 📫 Reach me at manavrameshkumar@gmail.com
Jun 2025 · 1 month
- Built a Python ETL pipeline to collect, transform, and structure unstructured web data into a queryable database.
- Applied TF-IDF NLP analysis on text corpora to deliver actionable strategic insights to the team.
Aug 2024 – May 2025 · 10 months
- Built interactive Power BI and Tableau dashboards with dynamic KPI visualisations and drill-downs by status, year, and geography — enabling data-driven decision-making across the organisation.
- Performed in-depth data analysis on 300,000+ records using SQL and Excel; conducted anomaly detection and validation to ensure data integrity.
- Integrated multi-source data (SQL + Excel) into a unified reporting layer, automating ETL workflows and reducing manual reporting effort by 20%.
- Identified that campaigns under 60 days had a 74% higher success rate — insight directly adopted by the strategy team.
- Presented weekly findings to a cross-functional team of 7, translating complex trends into clear, actionable recommendations.
Python · Streamlit · PostgreSQL · Groq API · Llama 3.3 (70B) · Plotly · 🔗 Live App
AI-powered sales analytics app that converts plain English into PostgreSQL queries — auto-renders Plotly charts across 50,000 records, 16 cities, and 8 product categories. Deployed live on Streamlit Community Cloud.
RAG LLM Groq API PostgreSQL Streamlit Plotly
Power BI · SQL · Excel
End-to-end interactive dashboard tracking $896K in funding across 300,000+ records. Full data cleaning, validation, and anomaly detection pipeline. Identified a 60% overall success rate and key campaign timing insights to guide strategy.
📊 Project Impact
| Metric | Result |
|---|---|
| Records analysed | 300,000+ |
| Funding tracked | $896K |
| Overall campaign success rate | 60% |
| Campaigns under 60 days success premium | 74% higher success rate |
Power BI SQL ETL Data Cleaning KPI Tracking
Python · Streamlit · Scikit-learn
Self-service ML tool for CSV datasets — automates ETL preprocessing, feature engineering, and model training with a clean multi-page UI, enabling non-technical stakeholders to run models independently.
AutoML Streamlit Feature Engineering Scikit-learn
Python · NLP · Scikit-learn
NLP pipeline classifying app review sentiment using TF-IDF + Logistic Regression / Naïve Bayes. Delivered actionable insights on user satisfaction trends with strong F1 scores.
NLP TF-IDF Logistic Regression Scikit-learn Seaborn
B.Sc Computer Science — SIES College of Arts, Science and Commerce, Mumbai · 2021–2024 · CGPA: 7.78/10
- 📜 Data Analytics Certification — Excelr Solutions (2025)
- 📜 Introduction to Generative AI — Coursera (2024)
- 📜 Introduction to Software Engineering — Coursera (2025)
Always eager to learn, build, and collaborate on exciting AI and data-driven projects!
