Looks like you have JavaScript disabled. For the full Twine experience, you will need to re-enable it.

I’m a full-stack AI engineer with hands-on experience building production AI systems, LLM agents, RAG pipelines, and agentic workflows. I focus on shipping reliable AI products across the full stack, from model training and synthetic data generation to MLOps and voice AI deployment. I’ve led end-to-end projects at Proventus Analytics and as a freelancer, architecting real-time voice agents, tool-using LLM workflows, and scalable AI services using FastAPI, Docker, AWS, and MLflow. I’m passionate about practical AI that delivers measurable impact for business automation and product development.…I’m a full-stack AI engineer with hands-on experience building production AI systems, LLM agents, RAG pipelines, and agentic workflows. I focus on shipping reliable AI products across the full stack, from model training and synthetic data generation to MLOps and voice AI deployment. I’ve led end-to-end projects at Proventus Analytics and as a freelancer, architecting real-time voice agents, tool-using LLM workflows, and scalable AI services using FastAPI, Docker, AWS, and MLflow. I’m passionate about practical AI that delivers measurable impact for business automation and product development.

Saroj Poudel

AI Engineer, Full Stack Developer, Data Scientist, +2





I’m a full-stack AI engineer with hands-on experience building production AI systems, LLM agents, RAG pipelines, and agentic workflows. I focus on shipping reliable AI products across the full stack, from model training and synthetic data generation to MLOps and voice AI deployment. I’ve led end-to-end projects at Proventus Analytics and as a freelancer, architecting real-time voice agents, tool-using LLM workflows, and scalable AI services using FastAPI, Docker, AWS, and MLflow. I’m passionate about practical AI that delivers measurable impact for business automation and product development.…I’m a full-stack AI engineer with hands-on experience building production AI systems, LLM agents, RAG pipelines, and agentic workflows. I focus on shipping reliable AI products across the full stack, from model training and synthetic data generation to MLOps and voice AI deployment. I’ve led end-to-end projects at Proventus Analytics and as a freelancer, architecting real-time voice agents, tool-using LLM workflows, and scalable AI services using FastAPI, Docker, AWS, and MLflow. I’m passionate about practical AI that delivers measurable impact for business automation and product development.

Available to hire

I’m a full-stack AI engineer with hands-on experience building production AI systems, LLM agents, RAG pipelines, and agentic workflows. I focus on shipping reliable AI products across the full stack, from model training and synthetic data generation to MLOps and voice AI deployment.

I’ve led end-to-end projects at Proventus Analytics and as a freelancer, architecting real-time voice agents, tool-using LLM workflows, and scalable AI services using FastAPI, Docker, AWS, and MLflow. I’m passionate about practical AI that delivers measurable impact for business automation and product development.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Intermediate

Intermediate

Language

English

Fluent

Nepali

Fluent

Work Experience

AI Engineer at Proventus Analytics

July 1, 2025 - June 30, 2026

Built real-time voice agents using OpenAI Real-Time API, Twilio, SIP, STT/TTS pipelines, and backend orchestration for work automation. Developed MCP-backed agent workflows with LangChain and CrewAI to connect LLMs with tools, APIs, and business systems. Replaced manual workflows with chat-based automation systems using retrieval, memory, tool calling through MCP, and API-driven execution. Integrated multiple platform APIs to build query-based knowledge bases for conversational search and internal support automation. Designed and deployed RAG pipelines with embeddings, vector search, document processing, and context-aware response generation. Built production AI services using FastAPI, Docker, AWS Lambda, S3, and cloud deployment workflows. Used MLflow for experiment tracking, model lifecycle management, evaluation logging, and reproducible AI/ML iterations.

AI Engineer at Freelance / Upwork

January 1, 2024 - December 31, 2025

Delivered AI automation tools, retrieval-based applications, and full-stack prototypes for practical business workflows. Implemented document processing, embeddings, vector search, workflow automation, and structured AI application logic. Deployed AI applications and prototypes using Docker, cloud services, and lightweight CI/CD-style workflows. Researched emerging AI frameworks and engineering patterns to select practical solutions for client requirements.