Navvya A.

Key Skills

Software

AWS SageMaker

Top Subject Matter

Policy and Contract Intelligence (RAG over enterprise documents)

Digital marketing analytics and ML-driven audience/creative optimization

E-commerce recommendations and event logging for model retraining

Top Data Types

Document

Video

Text

Top Task Types

Question Answering

Data Collection

Prompt + Response Writing (SFT)

Freelancer Overview

Policy & Contract Intelligence Assistant (RAG) | LangChain · ChromaDB · GPT-4o · FastAPI · Docker · RAGAS. Brings 2+ years of professional experience across legal operations, contract review, compliance, and structured analysis. Core strengths include ChromaDB, Python (pandas, and Matplotlib). Education includes Bachelor of Technology, Vellore Institute of Technology (2026). AI-training focus includes data types such as Document, Computer Code, and Programming and labeling workflows including Question Answering, Evaluation, and Rating.

Labeling Experience

Social Media AI Specialist Intern | QuantInsti Quantitative Learning

Prompt + Response Writing (SFT)

Built automated campaign performance pipelines to support AI/ML-driven optimization workflows for advertising. Applied ML-driven audience clustering logic to reduce manual segmentation effort for funnel targeting. Automated creative performance auditing and integrated findings into A/B testing frameworks for campaign iteration. • Tracked CPL, CTR, and CPA using Python-based extraction and analysis • Used audience clustering for TOFU/MOFU/MOFU funnel optimization • Automated creative auditing to identify high-converting ad variants • Supported A/B testing integration across Meta Ads and Amazon Ads

2026 - Present

Policy & Contract Intelligence Assistant (RAG) | LangChain · ChromaDB · GPT-4o · FastAPI · Docker · RAGAS

DocumentQuestion Answering

Developed a RAG pipeline to ingest enterprise PDF/DOCX documents and produce citation-grounded answers. Implemented retrieval with MMR and added hallucination guardrails for safer responses. Evaluated the system using RAGAS metrics such as faithfulness, relevancy, and context precision/recall. • Ingested and indexed documents into ChromaDB • Served answers via a FastAPI endpoint with retrieval constraints • Produced BA artifacts including requirements, user stories, architecture, and UAT plans • Ran RAGAS evaluation for quality assessment of generated responses

2022 - 2026

ML Engineering Intern | Nextfly Technology

DocumentData Collection

Instrumented and deployed an inference service for a recommendation module and captured structured event data for downstream training. Implemented collaborative filtering-based recommendations from large-scale user interaction logs. Set up a structured SQL schema to log real-time user events for model retraining and iteration. • Built SVD collaborative filtering recommendations using user interaction logs • Deployed a FastAPI inference endpoint and integrated it with the e-commerce backend • Captured click/add-to-cart/purchase events into a SQL schema • Enabled data-driven retraining workflows for recommendation improvement

2025 - 2025

Digital Marketing & Data Intern | ZombieApple

Automated weekly ad performance reporting and surfaced ROAS anomalies to support continuous optimization. Built ML-driven audience segmentation and predictive modeling to improve targeting decisions for digital advertising campaigns. Integrated insights into A/B testing frameworks for iterative performance improvements. • Trained a Random Forest model to predict creative fatigue using CTR decay and frequency features • Automated reporting using Python data extraction and visualization • Applied K-Means clustering for audience segmentation feeding DV360 custom audiences • Implemented threshold-based alerting to flag ROAS anomalies

2025 - 2025

Education

Work History