B.Tech, Computer Science & Engineering (Artificial Intelligence)
Noida Institute of Engineering & Technology  |  Batch 2022-26  |  CGPA: 8.63

   

Fresher Data Analyst with hands-on experience in SQL, Python, Power BI, and Excel. Built three end-to-end analytics projects covering payment fraud detection, customer churn segmentation, and workforce trend analysis; each grounded in real datasets with verified findings. Two months of industry experience at Ethara AI (Gurugram) evaluating LLM outputs across multiple models at scale.

Currently open to Data Analyst and Business Analyst roles across Noida, Gurugram, Bengaluru, Mumbai, and Chennai.

Languages & Querying

Visualization & BI

Libraries & Tools

Analytics Capabilities

Ethara AI  |  Gurugram  |  Feb 2026 to Apr 2026

• Rated 300+ LLM outputs on instruction following, truthfulness, prompt adherence, and verbosity across 5 project types including STEM reasoning at JEE Advanced level; daily output scaled from 20 to 50 tasks
• Completed 40 multi-model comparative evaluations across Gemini, GPT, Claude, Kimi, and GLM to identify failure points used for model improvement; team scaled from 15 to 50 members based on submission priority

• Sampled 55,554 transactions from a 6.3M-record PaySim dataset; confirmed 0 duplicates and 0 nulls in SQL and Python
• Found 18% of customers drove 62% of $8.97bn total transaction value
• Fraud concentrated in TRANSFER (0.94%, 49 cases) and CASH_OUT (0.33%, 51 cases); zero fraud in PAYMENT, CASH_IN, or DEBIT
• High-value transactions had 18% greater fraud probability (0.20% vs 0.17%); built a Power BI dashboard with 4 KPIs and an interactive slicer

• Built a 3-tier churn segmentation model on 8,000 users (25.89% overall churn) where single-metric models failed to separate groups
• Segmented into High Risk (492 users, 28.46%), Medium Risk (2,106, 26.54%), and Low Risk (5,402, 25.40%), each mapped to a retention action
• Found Student + Mobile the highest-churn combination at 29.92% through subscription x device cross-analysis; followed by Family + Mobile (27.82%) and Family + Desktop (27.52%)
• Skip rate was the strongest behavioral indicator: 0.3049 (churned) vs 0.2985 (retained)

• Cleaned 3,642 raw records down to 1,995 by removing 1,647 rows (45%) with missing layoff count and percentage
• US led with 256,559 layoffs; India second at 51,234; Post-IPO stage had the highest exposure at 204,132 layoffs
• 2022 was the peak year at 160,661 layoffs; Q1 2023 alone added 125,000+
• Consumer industry most impacted at ~45,182 layoffs; single largest event: 12,000 employees from one company