Gulzar A.
Mid-Level Data Scientist
Open to offersNew to PlatformAbout
Gulzar A. is a skilled Data Scientist hailing from Mumbai, India, with expertise in artificial intelligence and machine learning within the technology sector. Working remotely, Gulzar has made significant contributions to document question answering systems, particularly through designing and delivering a RAG system that enhances the querying of estimation and cost documents by end-users. He has proficiently engineered hybrid retrieval systems using pgvector and Elasticsearch to manage diverse document formats. With a focus on cost optimization, Gulzar developed a layer that routes queries across multiple models, significantly cutting costs. Additionally, he automated machine learning processes by setting up MLOps environments using AWS and Docker, coupled with performance monitoring via Grafana. Previously, Gulzar collaborated with a team in France, applying NLP techniques to survey data, enhancing client analytics through feature engineering and visual summaries. His formal education in Computer Science and Engineering from Jaipur Engineering College furnishes him with a robust foundation in AI specialization.
Experience
Data Scientist
MEP QS Services · 2025 — 2026Oversaw the architecture and implementation of a production RAG system for answering questions related to documents, enabling teams to query estimation and cost documents easily. Developed a hybrid retrieval pipeline integrating pgvector dense search with Elasticsearch BM25, combined using Reciprocal Rank Fusion to manage various document formats. Created strategies for chunking and embedding documents that maintain structure across diverse document types. Built a cost optimization layer for LLM with multi-model routing and semantic query caching to reduce expenses on queries. Established an evaluation loop for monitoring retrieval quality and costs using Grafana, with re-embedding of flagged chunks using Celery and MongoDB. Deployed production inference services on AWS utilizing FastAPI within Docker, alongside PostgreSQL and Redis.Junior Data Scientist
Insightrix · 2025 — 2025Collaborated with a senior data science team on various survey research initiatives, providing exploratory analysis, feature engineering, and modeling pipeline development in Python using Pandas and scikit-learn. Employed NLP techniques on open-ended survey responses for sentiment analysis and topic mining, creating visual summaries for client presentations.
Skills & Expertise
Education
- B.Tech, Computer Science and Engineering (AI Specialization)Jaipur Engineering College · 2021 — 2025
Interested in this professional?
Sign in as an employer to save this profile or invite Gulzar A. to a job.
Sign in as an employer