Lakshay Teotia

CS undergraduate specializing in Data Science

Faridabad, India 0+ yrs exp 85 · Excellent

About

CS undergraduate specializing in Data Science with hands-on GenAI experience (NHPC internship) and independently built agentic AI and deep learning projects.

Skills & Expertise (22)

Python Advanced

8.0/10

Years Exp

FastAPI Intermediate

7.5/10

Years Exp

rag systems Intermediate

7.0/10

Years Exp

ChromaDB Intermediate

7.0/10

Years Exp

Docker Intermediate

7.0/10

Years Exp

SQL Intermediate

6.5/10

Years Exp

Deep Learning Intermediate

6.5/10

Years Exp

AWS Intermediate

6.5/10

Years Exp

scikit-learn Intermediate

6.0/10

Years Exp

Data Structures & Algorithms Intermediate

6.0/10

Years Exp

OOP Intermediate

6.0/10

Years Exp

REST APIs Intermediate

6.0/10

Years Exp

Keras Intermediate

6.0/10

Years Exp

TensorFlow Intermediate

6.0/10

Years Exp

Transfer Learning Intermediate

6.0/10

Years Exp

Streamlit Intermediate

5.5/10

Years Exp

Groq API Intermediate

5.0/10

Years Exp

agentic AI Intermediate

5.0/10

Years Exp

TCP Intermediate

4.0/10

Years Exp

IP Intermediate

4.0/10

Years Exp

OSI Model Intermediate

4.0/10

Years Exp

VS Code

Work Experience

Gen AI Intern

NHPC Limited

May 2026 - Jul 2026

Designed and built RAG Studio, a production-ready multimodal Retrieval-Augmented Generation system enabling enterprise document intelligence over complex PDFs containing text, diagrams, and photographs. Engineered a three-pass self-healing text correction pipeline (rule-based regex → SymSpell dictionary lookup → contextual LLM repair) to eliminate OCR noise from scanned technical documents before vectorisation. Built a dual-path cross-modal retrieval architecture using BAAI/bge-large-en-v1.5 for text and CLIP ViT-L/14 for images, enabling text queries to surface relevant diagrams stored in a ChromaDB vector database. Delivered a fully containerised application via Docker, with a FastAPI async backend exposing SSE-streamed RAG responses with page-level citations, and a vanilla JS/HTML/CSS dark-theme frontend.