Experience

Postdot Technologies (Postman)

Bengaluru | Remote

Tools: OpenSearch, Python, JavaScript, SQL, Lambda, DynamoDB

Led design and optimization of search systems for Postman, delivering end-to-end solutions for discovery, relevance, and AI-powered retrieval.
Boosted engagement and metrics: improved search relevance and rankings (+1.5% usage, +3.2% click position, +1% MRR), and launched hybrid/semantic search (+1.8% engagement, +50% MRR for public APIs).
Engineered efficient, scalable systems: reduced vector search memory by 94% and latency by 80% via quantization, hybrid full-text and vector search, Elasticsearch custom routing, and data partitioning. Reduced query latency by 20%.
Enhanced specialized search: optimized code and URL search using advanced indexing, edge n-grams, FVH highlighting, synonym expansion, and wildcards.
Drove continuous improvement by building benchmarking and A/B testing frameworks, and refining ranking signals such as collaboration and multi-user interaction.
Mentored engineers and interviewed candidates.
Received the Ideaman Award and second place in a hackathon for innovative product features.

Aug 2020-March 2022
Mumbai, India

Tools: Python, JavaScript, SQL, MariaDB (MySQL), Redis, Frappe Framework

Created Frappe Wiki, an open-source web app suited for hosting dynamic web content, including version control, access control, and audit trail.
Redesigned Bank Reconciliation.
Started Dev Forums to facilitate knowledge sharing between developers.
Maintainer of Frappe Web Framework and Wiki. Contributor to ERPNext.

Jun 2018-July 2018
Mumbai, India

Worked on a project called Meaningful Insights, which involved predicting the probability of a deal closing. Around 800+ analysts use the software.
Extracted and cleaned the data.
Trained multiple models including Random Forest with 87% accuracy.