Meghana J.

Key Skills

Software

Other

Top Subject Matter

Meeting notes summarization and structured extraction

Mental health conversational AI

Visual-semantic product retrieval and recommendation

Top Data Types

Text

Image

Top Task Types

Text Summarization

Question Answering

Freelancer Overview

AI Meeting Notes Assistant (Python, Streamlit, Groq API, LangChain). Brings 1+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Other. Education includes Bachelor of Technology, G.B. Pant University of Agriculture and Technology (2025) and Intermediate, J.N.V. Nainital (2021). AI-training focus includes data types such as Text and Image and labeling workflows including Text Summarization, Question Answering, and Evaluation.

Labeling Experience

Data Science Intern (Digital Training Lab)

OtherText

As a data science intern, you cleaned and preprocessed datasets to improve data quality prior to model development. You performed exploratory data analysis (EDA) and built multiple interactive dashboards to communicate findings. You assisted in developing machine learning models to support prediction tasks, reaching 85% prediction accuracy. • Data preparation: cleaning and preprocessing with Pandas/SQL • Analysis: EDA and feature exploration • Modeling: Scikit-learn ML model development • Reporting: 5+ Power BI dashboards and accuracy target of 85%

2024 - 2024

Image-Based Product Recommender (Python, CLIP, FAISS, Streamlit)

OtherImage

You created an image-based product recommendation system using CLIP embeddings and vector retrieval. You indexed 1,000+ product images and evaluated visual-semantic match relevance during retrieval testing. You assessed retrieval performance using a reported ~92% match relevance metric. • Input: product images • Representation: CLIP embeddings • Retrieval mechanism: FAISS similarity search • Evaluation metric: ~92% match relevance in testing

2024 - 2024

Mental Health Check-In Agent (Python, LangChain, Groq API, Streamlit)

OtherTextQuestion Answering

You built an LLM-based mental health chatbot that generates conversational responses in a multi-turn setting. You tested the system on 100+ user queries to measure response behavior and relevance. You tuned conversational flow to improve both relevance and emotional tone in generated replies. • Input: user messages/questions about mental well-being • Output: chatbot responses with emotional-tone awareness • Performance: ~90% response relevance and emotional tone accuracy • Deployment: responsive UI with 1–2s average response time

2024 - 2024

AI Meeting Notes Assistant (Python, Streamlit, Groq API, LangChain)

OtherTextText Summarization

You summarized 50+ meeting transcripts into structured bullet points using LLM-driven summarization workflows. You evaluated summary quality and accuracy based on a reported >85% summary accuracy metric. You developed repeatable prompt patterns to support consistent output generation for multiple inputs. • Input content: meeting transcripts (unstructured text) • Output labels: structured bullet-point summaries • Quality metric: >85% summary accuracy • Testing: 10+ users for usability feedback

2024 - 2024

Education

Work History