Senior Data Scientist | MS CS Aspirant '26

Deepak Sharma

Bridging Engineering & Agentic AI.

Senior Data Scientist with 5+ years of experience building production-grade NLP and Generative AI pipelines. Expert in optimizing LLM workflows using LangGraph & AWS, reducing process latency by 60%. Currently focused on Model Context Protocol (MCP) and autonomous data systems.

Technical Arsenal

Agentic AI & GenAI

  • Agentic Patterns (ReAct, Plan-and-Execute)
  • LangGraph & LangChain Orchestration
  • RAG Pipelines (Hybrid Search & Chunking)
  • Model Context Protocol (MCP) & Tool Use
  • LLMOps (Governance & Evaluation)

Cloud Infrastructure (AWS/GCP)

  • AWS Bedrock & SageMaker (Production AI)
  • Serverless Architecture (Lambda, API Gateway)
  • GCP Vertex AI & BigQuery
  • Docker, Kubernetes & CI/CD Pipelines
  • Infrastructure as Code (Terraform/CloudFormation)

Data Engineering at Scale

  • Snowflake Cortex & Advanced SQL
  • Real-time Streaming (Kafka & KSQL)
  • Vector Databases (Qdrant, Milvus, Weaviate)
  • Search Engines (Elasticsearch, OpenSearch)
  • Data Lineage & ETL Automation

Core Machine Learning & NLP

  • Deep Learning (PyTorch, TensorFlow, Keras)
  • NLP & Transformers (BERT, GPT, T5)
  • Computer Vision (OCR/Tesseract, CNNs)
  • Recommendation Systems & Predictive Analytics
  • Python (Pandas, NumPy, Scikit-learn)
Professional Trajectory

Senior Data Scientist

@ Tiger Analytics

Mar 2023 - Present

Architecting Enterprise AI solutions. Designed Agentic AI systems using ReAct patterns and AWS Bedrock. Built a 'Scrum Assistant' reducing admin overhead by 50%.

Data Scientist

@ IntellectFaces, Inc.

Dec 2021 - Mar 2023

Led AI development for RytFit.ai. Reduced resume parsing time from 10s to <2s by optimizing BERT models. Built real-time streaming pipelines with Kafka.

Voice AI Engineer

@ MAKERDEMY

Feb 2020 - Dec 2021

Developed commercial Alexa Skills for US clients. Authored a technical course on Alexa Skill Development for 100+ students.

Key Projects

Enterprise Text-to-SQL Interface

Developed a natural language interface using Model Context Protocol (MCP) allowing dynamic schema retrieval for complex SQL databases.

GenAIMCPSQL

RytFit.ai Recruitment Engine

Automated recruitment platform featuring deep learning models for resume parsing, job calibration, and candidate ranking.

BERTNLPKafka

Agentic RAG Chatbot

Optimized chatbot pipeline for major insurance clients using Hybrid Search (Vector + Keyword) and LangGraph agents.

LangGraphAWSRAG

Thera Bank Loan Predictor

Predictive modeling system using Logistic Regression and Naïve Bayes to identify potential loan customers with 97% accuracy.

Scikit-LearnAnalyticsPython
Beyond the Code

Annapurna Base Camp (4,130m)

Dec 2025 - Jan 2026. A test of endurance and high-altitude planning. Trekking 80km through the Himalayas taught me that big systems—like big mountains—are conquered one calculated step at a time.

ResiliencePlanning
View Trek Journal