Jul 2025 – Oct 2025 · 3 mos · Calgary, Alberta, Canada · Remote

May 2024 – Apr 2025 · 11 mos · Calgary, Alberta, Canada · Remote

• Working on LLM based system for automated PR review

Aug 2023 – Apr 2024 · 8 mos · Calgary, Alberta, Canada · Remote

• Worked on improving Facebook Ads Quality using ML/NLP

Dec 2021 – Jul 2023 · 1 yr 7 mos

• Leading the efforts to implement LLMs like Flan, Bloom, Dolly , GPT etc using Langchain and Chroma ecosystem for long document QA, generation, extraction, and report generation.
• Working on various NLP and ML algorithms for Information extraction of visually rich documents and textual documents.

Sep 2017 – Nov 2021 · 4 yrs 2 mos

• 1. Created a retrieval-based system for extracting complex legal clauses with complex date resolutions, leading to a reduction of 20% of manual processing time on 60+ pages and faster review.
• 2. Lead the AI product development process for a Deep Document Understanding Platform, resulting in a reduction of deployment time from 1 month to 2 weeks by using a pluggable framework.
• 3. Engineered a contextual search retrieval system that decreased the property insurance contract processing time from 65 to 10 minutes.

Nov 2015 – Sep 2017 · 1 yr 10 mos · Mumbai Area, India

• 1. Devised a recommender system that suggests top 3 news articles daily to users based on their chosen professional interest topics.
• 2. Increased client engagement by 5% measured by interactions such as likes and comments.

Sep 2015 – Sep 2017 · 2 yrs · Mumbai Area, India

Apr 2015 – Sep 2015 · 5 mos · Mumbai Metropolitan Region · On-site

• Developed financial product recommendations tailored to customers of a banking client, resulting in a 20% increase in the quality of target customers based on their defined metric.

Sep 2013 – Oct 2014 · 1 yr 1 mo · Pune/Pimpri-Chinchwad Area · On-site

• Created a highly scalable and available NLP platform for interpreting textual documents, utilizing techniques such as shallow and deep parsing, coreference resolution, relationship extraction, named entity recognition (NER), topic modeling, sentiment analysis, and query interpretation.

Sep 2011 – Sep 2013 · 2 yrs · Pune/Pimpri-Chinchwad Area · On-site

• Implemented a data pipeline using a custom producer-consumer messaging queue and distributed indexing with Lucene and Solr to improve retrieval and semantic search on multiple long documents.

Aug 2010 – Aug 2011 · 1 yr

• Implemented ML-based content extraction to enhance search relevance of news articles on company website, leading to increased traffic and reduction in irrelevant content.

Jan 2010 – Sep 2010 · 8 mos · Mumbai Metropolitan Region

• "Keyword search in geo-spatial databases". - describes approach to support keyword queries, query interpretation, query translation into one or more SQL queries, and query execution and result ranking. It describes how geo-spatial ontologies can be used in understanding implicit relationships between query objects, and how R-tree based indexes can be effectively used for result ranking based on spatial relationships.