Google Health Partnership
Collaborated with Google to supply high quality, validated clinical content as training data for Google's health AI and LLM initiatives.
Each project is a story of collaboration, creativity, and craft. Here are some of our favorites.
Collaborated with Google to supply high quality, validated clinical content as training data for Google's health AI and LLM initiatives.
Managed the commercial and technical delivery of Tata 1MG's integration with Samsung Health, supplying structured drug data and clinical tools.
Collaborated with Elsevier to deliver 1,500 structured drug monographs via API integration.
Collaborated with Microsoft to supply structured drug content to Bing's health search results across four geographies.
Led the design of a retrieval augmented generation (RAG) pipeline to automate classification of drugs across six safety advisory categories.
Identified the absence of a systematic content quality measurement system and led the initiative to design and implement one.
Designed and delivered a state-of-the-art clinical interaction checker covering over 2 lakh drug-drug and drug-lifestyle interactions, now integrated into Tata 1MG's platform and licensed to partners.
Mapped 6,000+ drug salts against CDSCO's official approvals database to identify and remove unapproved substances from the Tata 1MG platform — resulting in the removal of ~300 non-compliant salts and ~5,000 associated medicines.
Integrated Tata 1MG's clinical content and real-time drug-drug interaction tool with Tata hospital EMR systems, providing doctors with in-workflow clinical decision support for anti-cancer drug prescribing.
Led the 0-to-1 build of Tata 1MG's e-diagnostics content platform — including PDP architecture, hierarchical content automation, SEO strategy, CMS schema, and lab nomenclature standardisation — driving 40% organic sales growth.
Spearheaded the transition from legacy siloed systems to a single unified Content Management System — consolidating cataloguing, content, and data management for drugs, diagnostics, and OTC into one cross-team platform.
Designed and implemented a three-tier ban management system — bringing overall ban compliance accuracy from 81% to 99%.
Mapped 500+ drugs to Schedule P storage temperature requirements — taking storage accuracy from 64% to 93%.
Designed and implemented a pin-code and state-level SKU restriction framework for 70 high-risk drug salts, blocking dispensing for ~5,000 medicines in regulated geographies.
Audited and corrected the chemical class and therapeutic class classifications for 3,000+ drug salts — improving classification accuracy from 65% to 95%.
Designed and built the production tracking system now used by Tata 1MG's YouTube team — enabling end-to-end visibility across shoot planning, production status, publication, and performance.
Served as project manager and individual contributor for Tata 1MG's COVID Task Force — designing the emergency vaccine distribution and order fulfillment channel for Government of India officials across Delhi NCR, serving 7,000+ individuals in two months.
Built an automated reporting system to identify and remove Not-for-Sale (NFS) medicines being listed on the platform — catching up to 200 non-compliant SKUs per month.
Designed the end-to-end governance framework for new drug salt additions to Tata 1MG's platform — embedding CDSCO approval verification, indication mapping, and compliance checks as mandatory gates.
Led a large-scale SKU-level data extraction and bulk update exercise for 60,000+ medicines — improving overall compliance accuracy to 95%.
Designed a strength-level content replication framework for paediatric drugs — enabling child-specific dosing, indications, and safety content to be scaled across 80 drugs and 20,000 SKUs.
Designed a new content and UI strategy for the VMS category — piloting on 40 SKUs and demonstrating measurable improvement in conversion, retention, and engagement.
Built an automated competitive intelligence pipeline using Python to scrape and compare Tata 1MG's drug and health content against national and international competitors.
Developed Tata 1MG's AI and automation strategy for the clinical content function — mapping the full content lifecycle and identifying AI insertion points to improve accuracy, efficiency, and scalability.
Led the scoping, content structuring, and product roadmap documentation for Tata 1MG's generics strategy — defining the content architecture and user experience model for generic drug discovery and substitution.
Designed and executed Tata 1MG's three-horizon search discoverability strategy — transitioning from traditional SEO to Answer Engine Optimisation (AEO) and Generative Engine Optimisation (GEO) for LLM-era AI answer surfaces.
Designed a multi-layer prescription validation system combining telemedicine review, image-level quality checks, and CDSCO schedule compliance — reducing fraudulent and non-compliant prescription orders.
Designed and delivered an AI-powered mapping engine linking drugs, diseases, and diagnostic tests — enabling clinically grounded cross-sell recommendations across Tata 1MG's pharma and diagnostics verticals.
Designed and executed a Reddit and community-platform growth strategy for Tata 1MG — building authentic health discussion presence, driving organic referral traffic, improving brand trust signals, and establishing community-sourced content as a GEO discovery asset.
Led Tata 1MG's Generative Engine Optimisation (GEO) initiative — restructuring drug, disease, and diagnostics content to align with how large language models retrieve, evaluate, and surface health information.
Led root-cause analysis of order return patterns at PharmEasy — identifying the top drivers of returns across product, operational, and fulfilment dimensions — and built a fraud detection and order lifecycle dashboard.
Analysed and optimised PharmEasy's checkout and delivery experience — identifying drop-off points across the order funnel, running targeted interventions, and improving checkout conversion and delivery process efficiency.
Owned end-to-end product management for Cyclegent's virtual cycling game — from 0-to-1 roadmap definition through to engagement loop design, retention mechanics, and analytics infrastructure — taking retention from 3% to 18%.