Shreya K.

Key Skills

Software

Other

Top Subject Matter

Employee sentiment analysis and social media/web analytics

LLM prompt engineering and prompt evaluation for a financial AI agent

Spatial analysis

Top Data Types

Text

Document

Top Task Types

Prompt + Response Writing (SFT)

Mapping

Classification

Tracking

Entity (NER) Classification

Freelancer Overview

GET – Data Analytics (Tata Power Mumbai). Brings 3+ years of professional experience across legal operations, contract review, compliance, and structured analysis. Core strengths include Power BI, Other, and QGIS. Education includes Bachelor of Technology, Sangam University (2025) and Higher Secondary Certificate (Class XII), Sophia Girls’ Sr. Sec. School (2021). AI-training focus includes data types such as Text, Geospatial, and Tiled Imagery and labeling workflows including Evaluation, Rating, and Prompt + Response Writing (SFT).

Labeling Experience

GET – Data Analytics (Tata Power Mumbai)

Text

Developed and deployed analytics-driven systems to interpret unstructured text inputs and surface structured insights for stakeholders. Built an employee sentiment analysis solution using Python and NLP to analyze large batches of textual queries and extract key themes for communication support. Ensured data quality across multiple sources to support reliable downstream reporting and decision making.• Analyzed 40+ Townhall queries to identify key sentiment themes.• Supported KPI tracking and engagement trend analysis using dashboard outputs.• Automated reporting workflows to improve insight turnaround time.• Validated and maintained data quality across 3+ data sources.

2026 - Present

AI Analyst (Contract) - Lean Research

TextClassificationTracking

I worked as an AI Analyst (Contract) focused on designing and optimizing prompt templates for a financial AI agent. I refined prompt logic to improve context retention and factual accuracy while reducing hallucinations. I also created an evaluation framework with test cases to ensure reproducibility and support agent scalability. • Architected and optimized prompt templates for LLM-based financial forecasting • Improved forecast accuracy by 17% and response consistency by 24% • Reduced hallucination by 30% via prompt logic refinements • Built a prompt evaluation framework with 50+ test cases for scalability

2025 - 2025

Jr. Data Scientist (Trainee) (Birla Institute of Scientific Research, Jaipur)

Mapping

Automated geospatial computations and produced analytical outputs for land parcel zoning using programmatic data processing. Engineered an exploratory data analysis pipeline and generated stakeholder-ready reports from spatial datasets. Performed spatial change analysis across thousands of parcels using geospatial libraries and exported actionable results for planning workflows.• Automated zoning calculations for 100K+ land parcels with improved accuracy and reduced runtime.• Executed spatial analysis on 2K+ parcels using GeoPandas and QGIS and identified 50+ changes.• Analyzed 4K+ real estate transactions and validated spatial data quality for urban planning.• Produced dynamic, bilingual stakeholder reporting outputs.

2025 - 2025

AI Analyst (Contract) (Lean Research, Remote)

OtherTextPrompt + Response Writing (SFT)

Architected and optimized prompt templates for a financial AI agent built with LLMs to improve prediction quality and response consistency. Refined prompt logic to retain relevant context and improve factual accuracy, reducing hallucinations. Created a prompt evaluation framework with a large set of test cases to measure and reproduce agent behavior across scenarios.• Improved forecast accuracy by 17% via prompt template optimization.• Reduced hallucinations by 30% through context retention and factuality refinements.• Defined a prompt evaluation framework using 50+ test cases.• Enhanced reproducibility and scalability of the AI agent.

2025 - 2025

Data Science Intern - Evoastra Ventures Pvt. Ltd.

TextClassification

I worked as a Data Science Intern supporting multiple machine learning projects and delivering analytics for product and customer insights. I developed and evaluated classification and prediction models using standard ML techniques and handled end-to-end workflow coordination. I also produced sentiment-based and pricing dashboards from large review datasets to support CX and strategic decision-making. • Co-led five ML projects to streamline workflows and reduce turnaround time • Built Gradient Boosting and KNN models for financial prediction and multi-class classification • Delivered sentiment analysis and pricing dashboards from 10K+ reviews • Supported collaboration across teams through shared model and reporting workflows

2024 - 2024

Show more

Education

Work History