Dec 2022 – Nov 2025 · 2 yrs 11 mos · Bengaluru, Karnataka, India

• Senior Product Manager - Risk & Decision Platform
• Designed and developed AI/ML-powered RTO Risk Intelligence platform enabling D2C merchants to predict and prevent return-to-origin losses.
• Designed and built India's first no-code workflow automation platform (KwikFlows) enabling D2C brands to build custom checkout flows and run A/B tests without engineering dependency. Validated product-market fit through 50+ merchant interviews achieving 100% adoption in target segment.
• Scaled merchant base from 70 to 2,500+ brands in 1 year with 60% product adoption from active clientele and 80% M3 retention rates. Generated $X ARR through tiered pricing model, with merchants creating 200+ workflows in first 40 days post-launch.
• Built MVP with 1 engineer and scaled to a self-serve diagnostic tool processing 4,500+ analyses monthly, saving merchants 20+ hours/week in manual operations. This platform has been adopted by 2,500+ brands across India's D2C ecosystem.
• Developed DIY RTO Risk Simulator allowing enterprise merchants to run real-time risk scenarios and optimize order acceptance strategies. Led team of 12 engineers and 5 data scientists to build and deploy ML models that reduced RTO rates across the merchant network.

Jan 2020 – Dec 2022 · 2 yrs 11 mos · Bengaluru Area, India

• Led end-to-end product development for fintech compliance and customer experience initiatives managing cross-functional team of 20+ members across engineering, data science, and operations.
• Designed and launched video-based KYC solution for digital account opening, increasing adoption by 26% and improving NPS by 5 points. This solution automated compliance workflows reducing manual verification time and improving customer onboarding experience.
• Built automated support platform using ML-based query classification and response generation. This system resolved 73% of customer queries automatically while maintaining 80%+ CSAT scores. Reduced partner issue resolution time from 72+ hours to <24 hours, improving CSAT by 14 points.
• Developed a community engagement platform for Niyo users, enabling peer-to-peer financial discussions and support. Scaled platform to 15K+ active members and 100K+ monthly active users, driving 65% improvement in user engagement metrics.

Jul 2019 – Jan 2020 · 6 mos · Mumbai, Maharashtra, India

• Built end-to-end augmented analytics system for media industry client extracting actionable insights from viewership data and content metadata. This system processed petabytes of streaming data to generate content performance metrics and audience behavior patterns.
• Developed and deployed text-based recommendation engine using NLP and collaborative filtering techniques to optimize content performance across multiple media properties. Delivered ML pipeline from data ingestion through model training to production deployment.

Jun 2018 – Jul 2018 · 1 mo · Gurgaon, India

• As a Data Analyst intern, My key responsibilities included :-
• → Designing and developing an API that fetched all the relevant information from restaurant listing platforms like Zomato and Swiggy to Limetray's database.
• → Brainstorming with internal stakeholders to define a rule-based system that qualified leads for restaurants and aligning them on the same.
• → Ensuring that the API was capable of generating new leads every week from multiple websites while adhering to the defined rule-based filtering criteria.
• → Implementing a config-based onboarding process that allowed the Sales team to manage the API with zero tech involvement.
• → Contributing to the overall business growth by streamlining the lead generation process, resulting in increased revenue and customer acquisition.

Feb 2018 – Nov 2018 · 9 mos · New Delhi Area, India · On-site

• As a Research Assistant at the MIDAS Lab under Prof Rajiv Ratn Shah (IIIT Delhi) and Prof Roger Zimmerman (NUS Singapore), I contributed to the following projects:
• → Speech Reconstruction using Multi-view Silent Video Feed: My team and I developed a CNN-LSTM-based Deep Learning System that utilized silent video feeds from multiple cameras recording the same subject to generate intelligent speech for a speaker.
• → Optimal Camera Placement: In addition to developing the Deep Learning System, I also identified the optimal placement of cameras that would lead to the maximum intelligibility of speech.
• → Publication: My paper on Speech Reconstruction was selected (6% Selection Rate) under "Brave New Ideas" at ACM Multimedia, 2018.

Jan 2018 – May 2018 · 4 mos · New Delhi Area, India · Remote

• Interned at an AI startup based out of New Delhi under Data Science Team. I worked on the following Projects.
• → Diabetic Retinopathy Detection- Automated the process of Classification of Diabetic Retinopathy using Multilayer Perceptron NeuralNetwork considering over 20 different attributes. Achieving the sensitivity of 74.3% and specificity of 82.3 %.
• → Epilepsy Detection System - Developed a Machine Learning based Epilepsy Detection system achieving accuracy of 95.6% along with a flask web application to predict seizure.

Aug 2017 – Jul 2018 · 11 mos · New Delhi Area, India

• Acting as Vice President of IEEE NSIT Student Branch chapters with the responsibility to manage all technical activities of the society while leading a team of more than 30 executive committee members.
• Delhi Section Congress (Oct 2017)
• → Prepared content, teaching material & delivering a weekly session to Freshers on Introduction to Data Science & Machine Learning.
• mentored students to apply these concepts to make useful industry standard projects.
• → Started a blog on the medium where notes were shared with students on a weekly basis.
• Java Bootcamp (Oct 2017)
• → Workshops conducted in partnership with Coding Ninjas, with over 150 participants.

Jun 2017 – Aug 2017 · 2 mos · New Delhi Area, India · Hybrid

• Worked under Dr Tanmoy Chakroborty on enlisted research projects :
• → Developed an algorithm which combines clustering and classification for ensemble learning.
• → My role was to Code & implement the algorithm wherein result from various algorithms were combined in form of Matrices & improved Classification results were obtained by Ensemble Model.
• URL Phishing Detection using Frequent Substring Patterns :
• → My role was to work on to finding frequent substring patterns to prepare an edge list of common patterns in Links.
• → Further, graphical inference was carried out to build a system for URL phishing detection