I'm an AI Engineer & Data Scientist building intelligent systems & scalable ML pipelines.
Scroll ↓
VT
About02
Notjust
anengineer.
Anarchitect
ofthought.
"I don't optimise models. I ask why the problem exists in the first place — then build something that makes it irrelevant."
Three years deep into AI — not as a spectator, but as someone who has shipped production systems, won hackathons, and consistently found the edge between what a model can do and what a product should do. Mumbai-based, globally minded.
0.0
GPA
0+
Years in AI
0+
Projects Built
0×
Hackathon Wins
Systems Thinker ✦ 01 ✦ Systems Thinker ✦ 01 ✦ Systems Thinker ✦ 01 ✦ Systems Thinker ✦ 01 ✦ Systems Thinker ✦ 01 ✦ Systems Thinker ✦ 01 ✦ Systems Thinker ✦ 01 ✦ Systems Thinker ✦ 01 ✦ Systems Thinker ✦ 01 ✦ Systems Thinker ✦ 01 ✦ Systems Thinker ✦ 01 ✦ Systems Thinker ✦ 01 ✦ Systems Thinker ✦ 01 ✦ Systems Thinker ✦ 01 ✦ Systems Thinker ✦ 01 ✦ Systems Thinker ✦ 01 ✦ Systems Thinker ✦ 01 ✦ Systems Thinker ✦ 01 ✦ Systems Thinker ✦ 01 ✦ Systems Thinker ✦ 01 ✦ Systems Thinker ✦ 01 ✦ Systems Thinker ✦ 01 ✦ Systems Thinker ✦ 01 ✦ Systems Thinker ✦ 01 ✦
Built & deployed CatalystOne — an enterprise AI chatbot on Hitachi hardware, supporting RAG, live web, and NL2SQL (CatalystQL) across 3 data sources.
Orchestrated the full app stack on Kubernetes with 100% on-appliance data residency and zero cloud dependency for air-gapped, enterprise-grade security.
Reduced data retrieval & query response times by ~60% by combining vector-based document retrieval with AI-generated SQL pipelines.
KubernetesRAGNL2SQLVector DBLLM
03
AI / RAG
35% Accuracy Boost
Data Science Intern
AGenC.ai· Nov 2024 – May 2025
Designed & deployed enterprise-grade RAG pipelines using LangChain & ChromaDB, improving retrieval accuracy by 35% for large-scale document processing.
Built a marketing analytics engine to detect low-performing products and trigger AI-driven content generation — boosting engagement by 40%.
Integrated custom embeddings for domain-specific semantic search, reducing irrelevant results by 25%.
RAGChromaDBLangChainAI AgentsEmbeddings
04
Full-Stack
20% Speed Gain
Python Developer Intern
Univ. of Texas at Dallas· Jun 2024 – Nov 2024
Built a Python/Flask rides & car-lending platform, improving database query performance by 20% through schema and controller optimization.
Designed and implemented scalable database architecture and core controllers, optimizing data retrieval speed by 20% for complex datasets.
Improved system scalability by refactoring key backend components and enhancing React UI functionalities.
FlaskReactPostgreSQLPython
Every role a new lens. Every problem a new language.
+25% institutional funding secured through strategic campaigns.
GDSC KJSIEIT
Marketing Expert
+50% event engagement through targeted community outreach.
"
Voices05
They Said
It.
"
Vyom is one of the most reliable team members to have. Curious, adaptive and most importantly taking the time to align with the outcome. Great work ethics and with potential to crack the code.
R
Rishabh Kapoor
Founder, AGenC.ai
"
Vyom has shown immense technical caliber and efficiency during the course of the internship. His immense knowledge and skills have proven to be a huge asset and have helped progress the project development cycle rapidly during the tenure of this internship.
S
Dr. Sumit Majumdar
Professor, University of Texas at Dallas
Connect
Initialize Transmission.
Drop a message onto the memory drive below. Let's build something real together.