Harshal Jhakal
Results-driven Data Analyst skilled in Python, SQL, Power BI, and Excel, with hands-on experience transforming complex datasets into actionable business insights. Built interactive dashboards and end- to-end analytics projects using real-world data to support data-driven decision-making.
Passionate about solving business problems through analytics, visualization, and continuous learning. Eager to leverage analytical thinking and technical skills to drive business growth.
Experience
Data AnalystFricon Systems
30% and improving memory efficiency by 15% across production workflows.
Designed and optimised ETL pipelines using Python (Pandas) and SQL, reducing data processing time by
Performed end-to-end data wrangling – cleaning, transforming, and validating raw datasets using Python
and Excel – improving data accuracy and consistency by 25%.
Implemented automated validation scripts to enforce data governance standards, reducing data quality
errors by 20%.
Executed SQL-based data extraction and aggregation across large relational databases, supporting efficient
reporting cycles.
Education
B.Tech CSE at Manav Rachna University(CGPA: 7.5)
B.Tech CSE at Manav Rachna University(CGPA: 7.5)
Class 10th at Vapi Public School(88%)
Class 10th at Vapi Public School(88%)
Skills
Projects
Zomato Restaurant Analysis
Built an end-to-end restaurant analytics pipeline in DuckDB/SQL and Python, analyzing 9,551 records across 15 countries, using window-style aggregates (FILTER, HAVING, CORR()) and a reusable named-query architecture shared between a Jupyter notebook and a Streamlit dashboard.
Uncovered and validated a non-obvious pricing insight — raw cost showed near-zero correlation with rating (r=0.08) while Zomato’s price tier showed a moderate correlation (r=0.40) — by identifying and correcting for a multi-currency data quality issue that would have invalidated a naive comparison.
Designed a parameterized SQL layer and interactive Streamlit dashboard (Plotly, cached DB connections) with live filtering across city and rating, translating raw exploratory findings into a stakeholder-ready tool.
E-Commerce Funnel Analysis
Built end-to-end e-commerce funnel analysis in PostgreSQL across 8,900+ events and 818 orders, uncovering a 2.4x conversion gap between desktop (25.6%) and mobile (10.5%) users using CTEs and conditional aggregation.
Analyzed channel-level conversion (social: 18.5% vs. referral: 14.8%) and 6-month cohort trends to identify high-leverage vs. low-ROI acquisition sources.
Designed and built an interactive HTML/Chart.js dashboard translating SQL findings into stakeholder-ready visuals (funnel, cohort trend, device breakdown).
Job Market Analysis Model
Designed and executed 5 complex SQL queries in PostgreSQL using JOINs, CTEs, and aggregations (COUNT, AVG, GROUP BY) to analyze a real-world job postings dataset and uncover hiring trends.
Conducted salary-to-skill correlation analysis identifying SQL (7,291 job postings) as the most in-demand skill and PySpark ($208K avg) as the highest-compensating, enabling data-driven career prioritization
Managed full project lifecycle using Git & GitHub for version control and portfolio publishing, producing actionable insights on optimal skills for data analyst roles.
certification
Infosys Springboard: Data Analysis with Python
Performed data cleaning, transformation, exploratory data analysis (EDA), and visualization using Python, Pandas, NumPy, and Matplotlib to derive meaningful insights from datasets.
Applied statistical analysis and data manipulation techniques to identify trends, summarize key findings, and support data-driven decision-making through clear visualizations and reports.
Forage: Data Labeling Job Simulation
Completed a Data Labeling Analyst job simulation by accurately classifying customer support messages based on Intent, Sentiment, and PII, ensuring consistent and privacy-compliant data annotation.
Reviewed ambiguous labeling cases, provided rationale for classification decisions, performed quality checks on annotations, and recommended improvements to labeling guidelines for greater consistency.
Udemy: ChatGPT for Data Science and Machine Learning
Leveraged ChatGPT to accelerate data science workflows, including Python code generation, debugging, exploratory data analysis (EDA), feature engineering, and model development.
Applied AI-assisted techniques to improve machine learning workflows, automate documentation, interpret model results, and optimize data preprocessing and analysis tasks.
Forage: Data Visualisation: Empowering Business with Effective Insights
Completed a simulation involving creating data visualizations for Tata Consultancy Services. Prepared questions for a meeting with client senior leadership.
Created visuals for data analysis to help executives with effective decision making.