Bhagwati Malav

Leading a team of engineers to build and scale a highly available, low-latency search platform handling 170M+ requests/day (P99: 350ms).
Focused on mentoring, coaching, and fostering a collaborative culture that emphasizes collective success.

1. Rebuilt the re-ranking platform into a configurable system supporting 40+ ranking models without requiring code changes.

2. Developed a scalable query understanding layer supporting 100+ attributes across product types, driving relevance across queries.

3. Driving operational efficiency: building a Search Agentic Tool to debug issues and help product/ops teams understand search behavior. Redefined test strategies to improve release quality and developer confidence.

4. Re-architected candidate retrieval to enable live vector search with dynamic switching between vector and lexical strategies based on runtime signals, optimizing for latency and cost.

5. Redesigned pagination flow with a unified system that dynamically fetches lexical, vector, and explore candidates to generate category-aware pages, supporting deep pagination.

6. Re-architecting ranking sublayer to optimize multiple business goals using a no-code, rule-based configuration setup.

7. Building an experience versioning and failure-trigger system that leverages DS signals to validate relevance, with automated rollback to stable versions.

8. Ensuring high availability via initiatives like index corruption handling, multi-AZ fault tolerance, and flow-specific rate limiting and isolation.

9. Led the initiative to introduce ScyllaDB at Zepto and implemented the first use case for search ranking. The system ingests multi-level signals (store-product, city-product, and store-level) to power ranking algorithms with sub-millisecond latency. Successfully implemented a high-performance solution supporting 500K+ requests per minute with exceptional read/write throughput.

• Leading a team of 6-8 engineers
• Building & leading multiple products in Search, Personalisation & Recommendation Space using NLP, LLM, AI.
• Building Paytm Super App Search Assistant (Supports vector search, api based retrieval, entity uid mapper, custom response mapper, text-to-sql, 70+intent/ entity classification)
• Deployed LLAMA2, 3 as a service/platform for internal use cases. Explored vllm, text generation inference etc., overall designed compute, response optimized solution.
• Built LLM + Deep Learning based multi lingual Intent Classification(70+ intents) and NER Detection for Indian Languages. Accuracy ~ 80% , 99th response time < 1 sec.
• Built a model to detect Indian languages from transliterated input. Accuracy ~ 80% - Built a model to detect Indian names based on search input.(at prefix level). Accuracy ~ 75 %
• Built in-memory Low Latency Trie based spellchecker which gives suggestion at prefix level, also uses phonetic sound, performance metrics, keyboard distance to give spell suggestions. 99.9 response < 20 ms

• Led a team of engineers and built multiple Search Products( ONDC Search, Federated Search Platform, Location Aware Search, Search As You Type, Autosuggestion etc.)
• Rearchitected Paytm Search Platform(Federated Search) to support vertical specific Search Experience.(No downtime, keeping 100% data consistency)
• Built Geo Hash based location filtered Search.
• Built ONDC Search/Discovery Platform (Location Aware Search)

• Built/Led the development of Paytm Search V2.0 experience from scratch. (Search As You Type)
• Built/set up elastic cluster from scratch, lot of benchmarking to get required SLA, making sure that we have highly available, fault-tolerant, highly scalable, elastic, low latency system. - Response_99.9 < 70 ms
• Lots of work around breaking existing Search Systems into multiple micro services with DDD(Domain Driven Design), separation of concerns, reusability, focusing on choosing the right tech stack etc.
• Re-Built existing search classifier to framework model which can run multiple versions of classifier without any code changes.

• Company Overview: Blibli(https://www.blibli.com/)
• Loved my journey at Coviam/Quinbay(Blibli.com), learned lot of things here :)
• Worked on building distributed, event driven, fault tolerant and highly scalable data auditing system which can be used across all the services.
• Blibli(https://www.blibli.com/)

• Company Overview: Blibli(https://www.blibli.com/)
• Worked on building Blibli.com B2B E-Commerce platform from scratch(Sales order creation, order approval, order creation, order fulfilment etc.) Worked with 6 other engineers, and led the technical direction of product.
• Worked on setting up all the required micro-services from scratch eg: order, customer, product, cart, finance etc, setup end to end data pipeline for the same.
• Blibli(https://www.blibli.com/)

• Joined as 6th member of tech team, worked on building Master Address system, Delivery Assist system. Python, Django, Postgres, aws lambda

• Company Overview: Blibli(https://www.blibli.com/)
• Worked on creating highly scalable Promo creation systems.(which touches multiple services eg: cart, search, product, pricing, merchant etc.)
• Worked on building Promo Scheduler, Email Scheduler.
• Worked on Price Scheduling, Price Indexing, Dynamic Pricing.
• Blibli(https://www.blibli.com/)

• Company Overview: Blibli.com
• Worked on adding new payment methods for various banks, worked on building order summary apis for merchant side users.
• Worked on building Birthday rewards feature which sends voucher to customers.
• Worked on building generic, distributed crawling framework which allows to extract products information for various product sources/ecommerce sites.
• Worked on building generic data extraction/massager framework.
• Worked on building real time price comparison application which works for various categories eg: handphone, computer, camera, health & food, drink, washing machine, tv, refrigerator.
• Worked on building distributed multi mapping recommendation feature which gives multiple master products recommendation for given merchant product.
• Blibli.com

• Worked on on various JD Edwards mobile IOS and android application issues.
• Working on Oracle hybrid mobile - MAF framework. Also handling AIS server issues which provides interface to the rest services for jde.

• Was part of Cerner Revenue Cycle team. Worked on building multiple features eg: appointment interaction etc. Java Rcp, Sql, Cerner command language.

• Worked on WPF and CCL.