Open to opportunities

Shubhayu Majumdar

@shubhayumajumdar

Message

Cloud Developer building production RAG and multimodal GenAI systems on Google Cloud.

India

Message

What I'm looking for

I’m looking to build reliable, production-grade GenAI and RAG features on Google Cloud—partnering with teams that value security, accessibility, and measurable quality so I can keep shipping impactful AI.

I’m a Cloud Developer who built production RAG systems and AI foundations end-to-end. At BlueVector AI, I architected a production RAG pipeline for the Iowa Department of Revenue, creating a retrieval corpus of 59,346 tax code and 91,338 administrative code extracts embedded into Vertex AI Datastore, with a self-validating reranking pipeline achieving near-perfect accuracy on DOR test cases.

I’m also the sole owner and engineer of AI Core Features, maintaining the full AI layer across Gemini and Vertex AI—multimodal generation, custom Vertex AI Datastore groundings, semantic search, and multiple RAG and GenAI pipelines across client projects. Beyond AI, I built a config-driven multi-database layer (MySQL, MongoDB, PostgreSQL), engineered MongoDB time series ingestion for 150,000+ records daily, and designed SnapApp Extensions to eliminate custom deployment overhead; I bring a security and compliance mindset from leading VAPT and WCAG remediation, and a research foundation from training a CNN+MLP model during my ML internship.

Experience

Work history, roles, and key accomplishments

Current

Cloud Developer

Current

BlueVector AI

Jul 2024 - Present (2 years)

Architected a production RAG pipeline for the Iowa Department of Revenue using Vertex AI Datastore, embedding 59,346 tax-code and 91,338 administrative excerpts to generate legally-cited determinations with near-perfect accuracy. Built company-wide AI core features across Gemini/Vertex AI (multimodal generation, semantic search, multiple RAG/GenAI pipelines) and scaled ingestion to reliably proces

RAG Data Pipelines Vertex AI Semantic Search Pub Sub Cross Database CRUD (MySQL MongoDB PostgreSQL)SnapApp Extensions

Associate Cloud Developer

BlueVector AI

May 2023 - Jun 2024 (1 year 1 month)

Integrated Google Document AI for bulk ingestion into SnapApp using custom and pre-built parsers, reducing import time by 90% for new parsers. Led app-wide VAPT and remediation by patching 8 critical SQLi/HTMLi vulnerabilities and resolving 200+ high/critical SonarQube findings, and delivered WCAG Level A/AA compliance for enterprise clients.

VAPT Security Testing SonarQube OWASP Style Threat Fixing WCAG Audits

Machine Learning Research Intern

Indian Institute of Technology Kharagpur

May 2022 - Aug 2022 (3 months)

Built a custom scraper to collect 35,600 user records and 73,000 images, then trained a two-branch CNN+MLP model to predict Instagram post popularity with 92% accuracy. Performed data preparation and model training to evaluate predictive performance on the collected dataset.

Web Scraping CNN+MLP Model data preprocessing Machine Learning Python Model Evaluation

Education

Degrees, certifications, and relevant coursework

Techno Main Saltlake

Bachelor of Technology, Computer Science (Artificial Intelligence and Machine Learning)

2020 - 2024

Grade: GPA: 8.86

Completed a B.Tech in Computer Science (Artificial Intelligence and Machine Learning) from Techno Main Saltlake from 2020 to 2024, earning a GPA of 8.86.