Open to opportunities

Rahul Goyal

@goyalrahul9649

Applied AI Engineer building production-grade RAG, semantic search, and forecasting systems for real-world industrial impact.

India

Message

What I'm looking for

I’m looking to build production RAG/LLM and time-series ML systems end-to-end—prioritizing retrieval quality, real-time inference, monitoring, and measurable impact. I want a team that values engineering rigor and deploys models to real workflows.

I’m an Applied AI Engineer with experience building production-grade ML systems at IIT Bombay, specializing in RAG pipelines, LLM-based document intelligence, and semantic search using Qwen and Ollama. I focus on end-to-end delivery—from data ingestion and retrieval quality to FastAPI backends and Streamlit interfaces.

In the HPCL project, I developed a RAG-based industrial knowledge assistant and engineered PDF ingestion, extraction, and section-aware chunking using pdfplumber. I’ve improved retrieval accuracy through hierarchical parsing, section indexing, and semantic search techniques, then shipped conversational access via Streamlit.

My data science work covers 400k+ time-series processing, optimization-based data reconciliation, neural forecasting, and anomaly detection for industrial readiness. I’ve also deployed models into enterprise production databases to TRL-8 readiness and automated adaptive model updates using MW-RBC techniques, with a consistent emphasis on reliability, monitoring, and measurable performance.

Experience

Work history, roles, and key accomplishments

Current

Applied AI Engineer

Current

IIT Bombay

Apr 2026 - Present (2 months)

Developing a Qwen/Ollama-based RAG industrial knowledge assistant for engineering documents. Built PDF ingestion, section-aware chunking, and a Streamlit conversational interface to improve retrieval for industrial repositories.

Qwen Ollama Semantic Search Pdfplumber Streamlit

Applied Data Scientist

IIT Bombay

Apr 2025 - Mar 2026 (11 months)

Processed 400k+ time-series samples and improved data reliability by 20% using optimization-based data reconciliation. Trained >80% accurate neural models for fouling prediction and built a FastAPI backend for real-time ML inference, monitoring, and what-if analysis.

Time Series Analysis Data Reconciliation Feature Engineering Neural Networks fastAPI Model Monitoring What If Analysis

Data Scientist - Analytics

IIT Bombay

Aug 2024 - Mar 2025 (7 months)

Preprocessed 1,500+ industrial variables and reduced false positives by 60% using advanced statistical analytics. Implemented PCA–SFA, Hotelling’s T, and anomaly detection for real-time fault monitoring, and supported TRL-8 production database deployment with adaptive model updates.

SFA Hotelling'S T Anomaly Detection Feature Engineering Model Evaluation

Project Lead, TIH-IoT

IIT Bombay

Apr 2023 - Mar 2024 (11 months)

Led a 4-member Chanakya Fellowship project team to simulate bioreactor operation, collect/analyze data, and derive research insights. Built a multi-variable controller and adaptive soft-sensors using hybrid modeling, Just-In-Time learning, and Gaussian Process Regression (GPR).

Bioreactor Simulation Multi Variable Control Soft Sensors Just In Time Learning Gaussian Process Regression Data Analysis Team Leadership