Utsav Paudel (also spelled Utsab Paudel, Utshab Paudel, Utsav Poudel) is a Machine Learning Engineer based in Kathmandu, Nepal, with 2.5+ years of experience building Generative AI systems, RAG pipelines, LLM applications, and Voice AI agents on AWS.

Who is a Machine Learning Engineer in Nepal?

Utsav Paudel is a Machine Learning Engineer in Nepal, currently working at Leapfrog Technology in Kathmandu. He specializes in Generative AI, RAG architectures, LLMs, and AWS-based AI solutions.

What does Utsav Paudel do?

Utsav Paudel builds production AI systems including RAG chatbots, Text-to-SQL engines, and Voice AI agents. He works with AWS Bedrock, Claude, LangChain, PyTorch, and FastAPI.

Utsav Paudel — Machine Learning Engineer | Generative AI, RAG & LLM Systems

01 / About

Bridging theory with
production-grade AI.

Machine Learning Engineer with 2.5+ years of experience in Generative AI, RAG architectures, and Cloud-Based AI solutions across Fintech and Healthcare sectors. I build systems that think, speak, and reason — from text-to-SQL chatbots to voice AI agents that schedule patients in real time.

Years Experience

Documents Processed

% LLM Accuracy

02 / Experience

Where I've built.

Machine Learning Engineer

May 2025 — Present

Leapfrog Technology Pvt. Ltd.

Visa Analytics Chatbot using AWS Bedrock + Claude Haiku 4.5 with a 4-stage RAG pipeline
GoLend AI: Text-to-SQL chatbot powered by Amazon Nova Pro + PostgreSQL pgvector
Voice AI Agent using Amazon Nova Sonic + Twilio, achieving 60% reduction in scheduling overhead
Automated patient referral pipeline using Textract + Nova Pro → OpenEMR via Lambda

Associate ML Engineer

Sep 2024 — May 2025

Fusemachines Pvt. Ltd.

Built RAG & GraphRAG systems with 40% accuracy improvement across 10,000+ financial documents
LLM extraction pipelines (GPT, Gemini) achieving 95% accuracy
Retrieval optimization delivering 50% query time reduction

Associate ML Engineer

Jan 2024 — Jun 2024

E.K. Solutions Pvt. Ltd.

Built 5+ scalable ML models/APIs using FastAPI + PyTorch handling 1,000+ daily requests
25% LLM accuracy improvement via Chain-of-Thought prompt engineering
35% transformer performance optimization with 20% memory reduction

ML Engineer Trainee

Oct 2023 — Jan 2024

E.K. Solutions Pvt. Ltd.

Foundational ML model development and training pipelines
Research and prototyping of NLP and computer vision solutions

03 / Projects

What I've shipped.

Project 01

Visa Analytics Chatbot

AWS Bedrock Knowledge Base + Claude Haiku 4.5 with a 4-stage RAG pipeline and ID-mapping citation strategy for accurate visa document retrieval and analysis.

AWS BedrockClaude Haiku 4.5RAGPython

0 Stage RAG Pipeline

Project 02

GoLend AI Assistant

SQL-RAG system using Amazon Nova Pro/Lite with BM25 + pgvector Hybrid Search, IVFFlat/HNSW indexing + MMR reranking, and AWS Guardrails for safe financial lending queries.

Amazon Nova PropgvectorBM25SQL-RAG

0 % Query Performance Gain

Project 03

Voice AI Patient Engagement

Real-time speech IVR using Amazon Nova Sonic + Twilio for automated patient scheduling. Integrated Textract + Nova Pro → OpenEMR pipeline for automated patient referral processing.

Nova SonicTwilioTextractOpenEMR

0 % Scheduling Overhead Cut

Project 04

Lender IQ

10,000+ document extraction platform with Neo4j GraphRAG, achieving massive query performance improvement for complex financial relationship queries.

Neo4jGraphRAGLangChainFastAPI

0 Documents Processed

04 / Skills

My toolkit.

Languages

PythonC++SQL

AI / ML Frameworks

PyTorchLangChainLangGraphLangSmithOpenCVGoogle ADKCrewAI

Infrastructure

DockerMLFlowRedisCeleryAlembic

Web Frameworks

FastAPIDjangoDjango REST

Databases

PostgreSQL (pgvector)Neo4jMongoDBSQLite

Cloud / AWS

AWS BedrockLambdaEC2SQSSNSGoogle Vertex AI

Utsav Paudel

Bridging theory withproduction-grade AI.