Mar 2024 – Jun 2025 · 1 yr 3 mos · Bengaluru, Karnataka, India · On-site

• Worked in the CX & Risk-Ops Analytics.
• Tools: SQL (65%), Tableau(20%), Sheets(10%), Python(5%)
• CX - (Aug'24 to Jun'25)
• 1. Work : Built funnels, visuals, and agent allocation models to optimise query routing, detect specific user demographics patterns, prioritise high-impact issues, tracking LLM performance through in-app analytics and human performance across chat, calls, and emails.
• 2. Impact : Improved resolution rate from 40% to 70% for our LLM chatbot, raised CSAT from 2.1 to 3.2, and increased SLA-resolved queries from 45% to 74%—achieving higher efficiency than our previous outsourced partner with 50% fewer agents.
• Risk Ops - (Mar'24-Jul'24)
• 1. Work : Built visualisations, funnel journey ,daily trackers, agent scorecards and data recon process setup between vendor and backend across 8+ Ops Queues.
• 2. Impact : strategic agent allocation, monitoring QA score of agents, capacity planning ,reduced backlog of 10k to 2k escalation tickets, increasing VKYC success rate by 20% by analysing failure reasons and flows.
• With the mentorship of Pushp Toshniwal and close collaboration with the entire CX team and Product Managers, I was able to ship impactful solutions daily. Together, we pulled the area out of a code red situation between September and December.

Jan 2024 – Mar 2024 · 2 mos

• 1) Processed and transformed historical data of customer information to build predictive ML models.
• 2) Employed statistical techniques to identify patterns and assess creditworthiness.
• 3) Classified customers into risk categories ("good" or "bad") for credit card approvals and limit setting.
• 4) Monitored and improved model performance by monitoring the evaluation metrics and made a AutoML app.
• 5) Through our constant teamwork and excellent mentorship from Gagandeep Uppal, we completed each and every deadline within the given timeframe to achieve our goal.

Jan 2024 – Mar 2024 · 2 mos

• ●​ Identified high-risk credit card customers (bad actors) using behavioral and repayment signals. ●​ Built target variables for new-to-credit customers using loan and delinquency data. Performed EDA and developed models optimized for precision, recall, and AUC.

Aug 2023 – Oct 2023 · 2 mos · India · Remote

• 1)The job was aimed to improve the quality and content of digital maps.
• 2)Through my detailing,research skills and knowledge of national and local geography,provided accuracy upto 90%.
• 3)Hence my work improved the overall experience of millions of users using digital maps in India.

Mar 2023 – Aug 2023 · 5 mos · Mumbai, Maharashtra, India · On-site

• Sucessfully analysed reviews for local and global brands for low to high end products in 5 different countries with people having different language and demographics.