Tanmay C.

Key Skills

Software

AWS SageMaker

Roboflow

Top Subject Matter

Retrieval-Augmented Generation for document Q&A

Customer churn prediction and segmentation using supervised ML

LLM-powered NLP

Top Data Types

Document

Image

Video

Top Task Types

Question Answering

Classification

Prompt + Response Writing (SFT)

Segmentation

Object Detection

Data Collection

Freelancer Overview

RAG-Based Document Q&A System (Personal Project, 2023). Brings 6+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include ChromaDB, LangChain, and OpenAI. Education includes Bachelor of Science, Louisiana State University (2020). AI-training focus includes data types such as Document, Computer Code, and Programming and labeling workflows including Question Answering, Classification, and Prompt + Response Writing (SFT).

Labeling Experience

Software Engineer (AI/ML Focus) — Deepwes Technologies (Jan 2021 – Present)

DocumentPrompt + Response Writing (SFT)

Integrated and operationalized large language models into internal tools and client-facing applications, which required preparing inputs/outputs for reliable model behavior in production. Built NLP modules for text classification, entity extraction, and semantic search that involved transforming unstructured text into structured representations. Designed REST APIs to serve ML/LLM capabilities for high-throughput inference workflows. • Integrated OpenAI GPT and Anthropic Claude into internal/client applications • Implemented text classification and entity extraction pipelines using Hugging Face Transformers and spaCy • Created embedding-based similarity retrieval workflows for RAG using pgvector/Chroma • Deployed ML model serving endpoints via FastAPI/Flask handling high-throughput inference

2021 - Present

RAG-Based Document Q&A System (Personal Project, 2023)

DocumentQuestion Answering

Developed an LLM-powered Retrieval-Augmented Generation (RAG) document Q&A system that required generating and managing retrieval-ready representations from user-provided documents. Implemented chunking strategies and metadata filtering to improve what content is surfaced to the model during Q&A. Exposed the solution via an API with a lightweight frontend to support iterative question answering over uploaded collections. • Built RAG pipelines using LangChain with ChromaDB and OpenAI embeddings • Applied chunking, metadata filtering, and re-ranking to enhance retrieval accuracy • Stored and retrieved embedding vectors for document collections • Integrated the RAG workflow into a FastAPI endpoint for interactive Q&A

2023 - 2023

Customer Segmentation & Churn Prediction (Deepwes Technologies, 2022)

Classification

Built an end-to-end machine learning pipeline for customer segmentation and churn prediction using CRM data to produce model outputs for downstream business decisions. Managed automated monthly retraining so the model predictions could be refreshed as new data arrived. Delivered segment-level insights through analytics reporting tied to the churn model results. • Developed ML models using Scikit-learn and XGBoost for churn prediction • Automated data preparation and model retraining using Apache Airflow • Produced segment-level insights using Pandas-based reporting • Worked with a labeled dataset implied by churn/CRM outcomes for supervised prediction tasks

2022 - 2022

Education

Work History