Selected work

Projects

From predictive models to working AI applications.

01 / Predictive modeling & optimizationFIFS Gameathon 2025 2.0 · 1st Place

Fantasy Cricket Prediction & Team Optimization

A fantasy cricket app that helps users choose balanced teams based on player form, match conditions, and predicted performance.

Won 1st Place at FIFS Sports Data Gameathon 2025 2.0.
Engineered player-form, venue, and matchup features using pandas and NumPy for IPL fantasy-point prediction with PyCaret models.
Implemented PuLP integer programming to optimize 11-player teams within 100 credits and 15-player squads within 145 credits; deployed a Streamlit app with API-based data ingestion and CSV export.

PythonpandasNumPyPyCaretPuLPStreamlit

GitHub repository ↗

Live demo ↗View certificate ↗

02 / LLM orchestration & LaTeX toolingSection routing · compilation repair

AutoCV

An AI-assisted resume editor that helps users refine their content, review changes, and create a polished PDF.

Routes editing instructions to specific resume sections to preserve unrelated LaTeX content.
Compiles candidate PDFs and exposes proposed changes for human-in-the-loop review.
Integrates Gemini, Groq, and Mistral APIs for contextual edits and compilation-error repair.

PythonStreamlitLaTeXLLMs

GitHub repository ↗

03 / Grounded generation & workflow orchestrationSource ingestion · grounded generation

SourceSpark

A planning workspace that turns source documents into grounded campaign ideas, with tools to review, organize, and schedule content.

Ingests PDF, DOCX, XLSX, and CSV content into a unified planning workflow.
Validates generated ideas against source evidence and tracks approval decisions before scheduling.
Persists reusable topic libraries and portable workspaces with Markdown and CSV serialization.

PythonStreamlitGeminiGroq

GitHub repository ↗

04 / Document intelligenceOCR · classification · extraction

DocXtract

A document-processing app that turns scanned files into readable text, useful summaries, and organized information.

Extracts text from scanned images and PDFs with Tesseract OCR.
Integrates Gemini and Groq for document classification and structured field extraction.
Exposes extracted attributes, document summaries, and serializable JSON results through Streamlit.

PythonTesseract OCRStreamlitLLMs

GitHub repository ↗

05 / Conversational analyticsAST validation · configurable 5-second timeout

DataBuddy – LLM-Powered Conversational Data Analyst

A conversational data analyst that lets users explore spreadsheets, ask follow-up questions, and understand results through charts.

Built a Streamlit analytics assistant that converts natural-language queries into pandas code for CSV/Excel analysis and interactive Plotly visualizations.
Implemented dataset profiling, AST-based code validation, and subprocess execution with a configurable 5-second timeout.
Integrated Groq, Mistral, and Gemini APIs for LLM routing, provider fallback, and conversation summarization to support contextual follow-up queries.

PythonpandasStreamlitPlotlyGroqMistralGemini

GitHub repository ↗

Live demo ↗

06 / NLP & explainable AIBERT · 0.8830 weighted F1

NLP-Based Automated Fake News Detection

A news-claim classification project that explores how language and credibility signals can help assess whether a claim is trustworthy.

Developed a LIAR2 classification pipeline using claim text, fact-checker justifications, and speaker credibility, with 18,369 training and 2,296 test examples.
Fine-tuned bert-base-uncased with a hybrid text–credibility classifier in PyTorch, achieving 0.8830 weighted F1 versus 0.8684 for SVM.
Benchmarked SVM, Random Forest, and BiLSTM; engineered 50,000 TF-IDF features and applied SHAP to explain SVM predictions across 200 test samples.

PyTorchBERTScikit-learnTF-IDFSHAP

GitHub repository ↗

07 / Computer vision & deep learningPatch embeddings · self-attention

PneumoVision ViT

A chest X-ray classification app that explores deep learning for identifying signs of pneumonia, with an interface for uploading images and viewing predictions.

Implements pre-LayerNorm transformer blocks with residual connections and a GELU-based MLP classification head.
Trains with AdamW and cross-entropy loss, validation-F1 checkpointing, and early stopping.
Exposes image-upload inference through a Streamlit interface backed by the trained PyTorch model.

PyTorchVision TransformerAdamWStreamlit

GitHub repository ↗