Hemang Goyal

Senior Data Scientist | AI/ML Systems Builder

Building scalable AI systems that bridge cutting-edge research with real-world business impact. 8+ years of experience in LLMs, AI Agents, NLP, and MLOps across finance and technology.

8+Years Experience
20+ML Projects
MillionsUsers Impacted

About Me

Senior Data Scientist & AI Systems Architect

Background

I'm a Senior Data Scientist with expertise in building and deploying scalable Machine Learning and Generative AI systems. My career spans finance and technology, where I've designed enterprise-grade AI platforms, optimized ML pipelines, and mentored teams in data science best practices.

Philosophy

I believe in bridging cutting-edge AI research with practical business outcomes. My focus is on building systems that are not just technically sound, but also deliver measurable impact at scale. Transparency, scalability, and user-centricity drive my work.

Education

B.Tech in Electrical and Electronics Engineering

Indian Institute of Technology (IIT) Guwahati | 2013–2017

Impact Metrics

Measurable Results at Scale

8,000+

Users Adopted IVYGPT

Enterprise GenAI platform deployed globally

80%

QA Time Saved

AI Agent evaluation framework reduced manual testing

40%

Cost Savings

R-CNN table extraction automation across 100K+ PDFs

2 weeks → 10 min

Process Acceleration

95% accurate commentary generation pipeline

Career

Professional Experience

Senior Data Scientist

Invesco

Hyderabad, India

Apr 2021 – Present

  • •Built IVYGPT, enterprise GenAI platform with LLMs and AI Agents (8,000+ users)
  • •Designed AI Agent evaluation framework (LangSmith, RAGAS) reducing QA by 80%
  • •Developed commentary generator pipeline with 95% accuracy (2 weeks → 10 min)
  • •Mentored 20+ professionals in Citizen Data Scientist program
  • •Fine-tuned DistilBERT for ESG classification (92% accuracy, 0.94 AUC)
LLMs
AI Agents
RAG
Langchain
NLP
Langgraph
SQL
AWS
Azure
LangSmith
Python

Data Scientist

Decimal Point Analytics

Mumbai, India

Aug 2019 – Mar 2021

  • •Developed R-CNN-based table extraction processing 100K+ PDFs
  • •Automated AWS workflows reducing costs by 40%
  • •Built news recommendation models with >90% accuracy
  • •Saved 100+ hours/month through automation
Computer Vision
R-CNN
AWS
Python
Deep Learning

Software Engineer

Samsung

Noida, India

Jun 2017 – Jul 2019

  • •Led R&D for Android Sound Quality (Dolby ATMOS, SoundAlive)
  • •Deployed AI tools to millions of Samsung devices globally
  • •Built context-aware music recommender using hybrid filtering
  • •Optimized audio processing algorithms for mobile devices
Audio Processing
Mobile AI
Recommendation Systems
C++
Python

Technical Expertise

Skills & Technologies

Languages & Tools

Python, SQL, TypeScript, AWS, Azure, GCP, Airflow, Docker, FastAPI, Pinecone, Snowflake, Databricks

ML Frameworks

PyTorch, TensorFlow, Hugging Face, LangChain, LangGraph, Neo4j, PySpark, Pydantic, DSPy, CrewAI, LlamaIndex

GenAI & LLMs

LLMs (Llama, Mistral, Claude, Gemini), AWS Bedrock, GCP Vertex AI, AI Agents, RAG, AutoGen, MCP, Google ADK, Multimodal AI, Fine-tuning (LoRA, RLHF), Prompt Engineering

MLOps & DevOps

CI/CD, MLflow, SageMaker, Kubernetes, Terraform, A/B Testing, Model Monitoring, ML Pipelines, Grafana, Streamlit

Portfolio

Featured Projects

IVYGPT

Enterprise GenAI Platform

Internal AI platform powered by LLMs and AI Agents for RAG-based document QnA, interrogation, and multilingual translation.

8,000+ global usersReal-time processing
LLMs
AI Agents
RAG
LangSmith
Python
Databricks

Model Commentary Generator

Automated Report Generation

LLM-powered pipeline for automated financial commentary generation, reducing manual reporting time from 2 weeks to 10 minutes.

95% accuracy500+ documentsDaily automation
LLMs
NLP
Pipeline Architecture
Batch Processing
Python

R-CNN Table Extractor

PDF Data Extraction at Scale

Computer vision-based solution for extracting tables from PDFs, processing 100K+ files across 5,000 companies.

100K+ PDFs processed40% cost savingsAWS automation
Computer Vision
R-CNN
AWS
Deep Learning

Cloud Cost Forecasting Tool

Cost Optimization Platform

ML-powered forecasting tool for predicting and optimizing cloud infrastructure costs with predictive analytics.

20% accuracy improvementReal-time insightsCost reduction
ML
Time Series
Analytics
Python

Credentials

Certifications & Continuous Learning

Neural Networks and Deep Learning

Coursera • 2019

Deep Learning
Neural Networks

Machine Learning Engineer Nanodegree

Udacity • 2020

ML Engineering
MLOps

AI for Trading Nanodegree

Udacity • 2020

AI
Trading
Algorithms

ChatGPT Prompt Engineering for Developers

DeepLearning.AI • 2024

LLMs
Prompt Engineering

AI Agents in LangGraph

DeepLearning.AI • 2024

AI Agents
LangGraph

MCP: Build Rich-Context AI Apps with Anthropic

DeepLearning.AI • 2025

MCP
AI Applications

How Diffusion Models Work

DeepLearning.AI • 2025

Generative AI
Diffusion

Get In Touch

Let's Connect

Interested in collaborating or discussing AI/ML opportunities?

Send a Message