Feb 2022 – Apr 2022 · 2 mos · Dubai, United Arab Emirates

• A 12-week immersive program consisting of a consulting engagement with FAST, an innovative foundation based in Armenia, solving one of their strategic challenges to help build Armenia as a centre for innovation in AI.

Aug 2020 – Aug 2021 · 1 yr · Bengaluru, Karnataka, India · On-site

• Constructed predictive models for 300+ user attributes using customer browsing and purchase data, achieving coverage for more than 20% of the Japanese population with an accuracy rate exceeding 80%.
• Developed a scalable Data Science Pipeline automating end-to-end data management, model training, performance evaluation, prediction monitoring, and deployment workflows. This resulted in an over 80% reduction in new model development time.
• Conceptualized distinctive features for behavioural marketing, categorizing users into interest-based segments by analyzing their past browsing and spending behaviour. Successfully implemented proof of concept (POC) for two of Japan's top IPs.

Aug 2020 – Aug 2021 · 1 yr

• Built predictive models for 300+ attributes with 80%+ accuracy across 20M+ users (~20% of Japan’s population).
• Developed automated ML pipeline covering ingestion, training, evaluation, deployment, cutting cycle by 50%.
• Delivered behavioral segmentation tools for marketing, boosting engagement for Japan’s top IPs.

Sep 2019 – Aug 2020 · 11 mos

• Managed a team of 5 Data Scientists and Product Engineers to model various Analytical solutions for Retailers
• Developed a Visual Business Intelligence tool for allowing Retails to track Sales and Stock Transactions across various channels. Provides real-time analytics of metrics such as Rate of Sales, Sell Through, and intelligent triggers for important events such as store stockouts and missed opportunities.
• Integrated the BI tool at either on-prem or cloud setups with 20+ clients with 600+ active users with the Average Customer Value per user being $4000

Feb 2018 – Sep 2019 · 1 yr 7 mos

• Developed a Visual Search Engine for faster discovery of similar items. The search engine can execute a query from amongst 100 Million images and return similar images in under 500ms
• Built a deep-learning-based feature extractor to cluster images based on visual and semantic similarity; decreasing user's cognitive load of going through over 60k images to 80-100 themes of similar Style

Nov 2017 – Jan 2018 · 2 mos

Feb 2018 – Aug 2020 · 2 yrs 6 mos

• Led a 5-member data science team building analytics solutions that influenced buying decisions across 30+ retail clients.
• Built a real-time Visual BI tool for sales, stockout alerts, and KPIs, adopted by 600+ users across 20+ enterprise accounts.
• Secured $4,000 average ARR per customer by integrating BI platform across hybrid environments.

Jan 2018 – Feb 2018 · 1 mo · Bengaluru, Karnataka, India

Jun 2017 – Dec 2017 · 6 mos · Bangaon, West Bengal, India

Aug 2016 – Dec 2016 · 4 mos · Bangalore

• Shifting the whole back-end system from wordpress to a tailored made system. Along with providing new features.

Jul 2016 – Jul 2016 · 0 mo · Bangalore

• Web-Development Intern