Data & Analytics

Drafts application emails or LinkedIn messages for specific job postings, leveraging Rahul Kotian's resume and strongly emphasizing his research paper on Spoken Language Understanding.

General SOP for common requests related to data, radar, self.

Provides detailed, step-by-step explanations and visualizations of computer science algorithms (especially sorting) and complexity analysis using plain text without LaTeX formatting.

Generates a comprehensive set of evaluation metrics and visualizations for classification models, including classification reports, confusion matrices, ROC curves (binary and multi-class One-vs-Rest), and density plots of predicted probabilities.

Load audio files from a directory, parse labels from filenames, generate random VAD segments, extract STFT features (mean along axis 1, converted to dB), and split the dataset into train/test sets.

Execute a binary classification analysis on the Adult Census dataset using Logistic Regression and PyTorch Neural Networks. Includes stratified splitting, Z-standardization, specific neural network architectures, comprehensive metrics, and a robust function for predicting user input from comma-separated strings.

A Python program to detect anomalies in videos using the VideoMAEForPreTraining model. It processes videos by dividing them into 16-frame clips, extracts embeddings using an unmasked boolean mask, and compares them against a normal behavior profile using Mean Squared Error (MSE).

Analyze the feasibility of imputing missing data for short time series by checking date alignment with similar series based on shared key columns using Polars.

Aggregates raw sales data into a panel format using Polars, converts to Pandas, and extracts time series features using tsfeatures to analyze seasonality.

Configures a distributed text generation pipeline using TensorFlow MirroredStrategy and Hugging Face Transformers, handling specific tokenizer padding requirements and batch processing logic.

Performs deep, expert-level analysis of technical concepts (specifically AI/ML architectures) by decomposing them into components, associating them with existing research/theory, evaluating pros/cons and readiness, and synthesizing a final conclusion.

Executes a comprehensive time series forecasting pipeline using StatsForecast and Polars, featuring 52-week seasonality, specific cross-validation parameters (h=5, n_windows=10), loop-safe ensemble aggregation, WMAPE calculation, non-negative constraints (including intervals), visualization, and ID splitting.

Designs sales compensation models for Series-A subscription-based enterprise software startups using a 6-step Monte Carlo simulation methodology. Integrates advanced analytics like predictive modeling and CLTV analysis to generate executive-level documentation such as ATS-optimized resume bullet points and interview responses.

Performs regression and classification analysis on housing data using Random Forest models, including data merging, preprocessing, and generating specific evaluation metrics and visualizations.

Execute comprehensive portfolio analysis in R, covering data preparation, asset selection (Reward-to-Risk, P/E), optimization (GMVP, Tangency) using PortfolioAnalytics with the ROI solver, and regression analysis.

Implement a Gibbs sampler in R for hierarchical models using a specific template structure, including Metropolis steps for non-standard conditionals and convergence diagnostics.

Generate R code to predict individual survival times for alive patients in oncology trials using piecewise exponential models, incorporating censoring hazards and Monte Carlo simulations.

Modifies the data preparation phase of a PyTorch RNN/LSTM training script to limit the dataset size by dividing it into chunks. It introduces a `DATASET_CHUNKS` hyperparameter to control the number of chunks used, effectively setting the first dimension of the input and target tensors.

Implement a PyTorch script to generate synthetic linear equation data (ax + b = c), train and compare Mixture of Experts (LSTM and Transformer) against Single General Models (LSTM and Transformer), and visualize the training loss comparison.

Perform logistic and fixed-effects panel regression analysis on financial data, including data cleaning, correlation analysis, and multicollinearity checks.

Generates a complete Python project featuring a Tkinter GUI and a Streamlit web dashboard for a bar stock exchange system. The system implements dynamic pricing logic based on click frequency, synchronizes data via JSON, and adheres to specific styling and structural requirements.

Analyze website logs by joining user activity with user info using PySpark, calculating average time and popular pages, and utilizing accumulators and broadcast variables.

Builds a binary classification neural network for the Adult Census dataset using robust, dynamic preprocessing. Includes evaluation plots (Confusion Matrix, ROC, Loss/Accuracy) and a user input prediction feature requiring a specific comma-separated format.

Generates H2-formatted content for vendor profiles and shipping/refund policies based on provided raw data, statistics, and descriptions, ensuring a third-person perspective.