3. Data, Science & AI

Guides Raman spectrum peak fitting with Lorentzian/Gaussian models, emphasizing data verification and physical constraints for material analysis.

Guides minimal GPT-2 inference implementation including checkpoint parsing, BPE tokenization, and avoiding common pitfalls in constrained environments.

Enables programmatic Excel spreadsheet operations including data analysis, pivot tables, and multi-sheet workbook management using pandas and openpyxl.

Generates structured CSV summaries from log files by aggregating counts across date ranges and severity levels (ERROR, WARNING, INFO).

Analyzes geospatial data with geopandas, handling coordinate projections for distance calculations, spatial filtering, and earthquake data processing.

Optimizes LLM inference request batching and scheduling to minimize cost while meeting latency and padding constraints through strategic grouping and shape management.

Provides expert guidance for optimizing slow SQL queries, improving database performance through query plan analysis and benchmarking strategies.

Guides Bayesian Network tasks including structure learning, parameter fitting, causal interventions, and sampling using pgmpy or similar libraries.

Processes CSV data for sensor/simulation results, including time-series analysis and missing value handling with pandas.

Provides technical guidance for DNA assembly methods including Golden Gate, Gibson, and restriction enzyme cloning, with focus on primer design and fragment assembly strategies.

Extracts weight matrices from black-box two-layer ReLU neural networks using input-output queries for model extraction and reverse engineering.

Provides best practices and workflows for analyzing exoplanet light curve data, including detection methods and troubleshooting.

Guides migration of legacy Python 2 scientific computing code to Python 3 with modern libraries for data processing and analysis.

Fits first-order dynamic models to step response data, extracting gain (K) and time constant (tau) parameters for system analysis.

Simulates vehicle motion dynamics, calculates safe following distances and time-to-collision, and implements cruise control state machines for automotive systems.

Downloads real-time or historical streamflow data from USGS using the dataretrieval package for hydrological analysis.

Performs dimensionality reduction on multivariate data using Principal Component Analysis with varimax rotation to identify underlying factors and reduce variable collinearity.

Searches a bundled flight dataset to find available routes and dates for proposal or verification.

Provides automated planning utilities for PDDL domains and problems, including plan generation, validation, and output saving using classical planners.

Guides implementation of tensor parallelism in PyTorch for distributed training, including sharding linear layers across GPUs and simulating collective operations.

Guides deployment of HuggingFace transformer models using Flask APIs with model caching and REST endpoint configuration.

Provides theorem proving assistance in Lean 4, including error resolution, mathlib lemma search, and compiler-guided proof repair.

Extracts tables from PDF files and converts them into pandas DataFrames for efficient data processing and analysis.

Linearizes nonlinear dynamic systems around operating points to enable control system design and analysis.