3. Data, Science & AI

Enables LLM reasoning and action loops through LangChain ReAct agent implementation with tool binding for AI workflows.

Expert integration of NVIDIA cuBLAS and cuDNN libraries for GPU-accelerated linear algebra, deep learning layers, and mixed-precision operations.

Processes 10X Genomics single-cell RNA-seq data using Cell Ranger for alignment, quantification, and analysis.

Provides techniques to minimize simulation-to-reality gap in AI models, ensuring effective transfer to physical environments through validation methods.

Enables interactive visualization and review of genomic data using the Integrative Genomics Viewer (IGV) for bioinformatics analysis.

Analyzes SEM-EDS data to determine material morphology and elemental composition for scientific research and materials characterization.

Optimizes CUDA Unified Memory configurations, prefetching strategies, and migration profiling for GPU-accelerated AI applications.

Automates hyperparameter tuning for ML models using Optuna with advanced search, pruning, and visualization capabilities.

Constructs tight-binding models using maximally localized Wannier functions for electronic structure calculations in condensed matter physics.

Provides high-accuracy SNV and indel detection using DeepVariant's deep learning model on genomic sequencing data.

Manages feature stores for consistent online/offline feature serving and training-serving parity in ML workflows.

Solves scheduling, configuration, and assignment problems using constraint programming techniques in AI workflows.

Enables distributed machine learning training, hyperparameter search, and resource management using the Ray framework.

Expertise in designing and developing battery management system algorithms, including validation and computational modeling for battery systems.

Generates navigation meshes and handles pathfinding for game AI across Unity, Unreal, and Godot, supporting dynamic obstacles and agent configuration.

Analyzes high-energy physics data using ROOT framework for histogramming and statistical processing.

Provides schema design, migration management, relation handling, and query optimization for Prisma ORM database integration.

Analyzes nanoparticle dispersion stability through zeta potential measurements, aggregation kinetics modeling, and shelf-life prediction.

Generates and counts combinatorial objects including permutations and combinations for mathematical and computational tasks.

Provides time series forecasting models for predicting business metrics and optimizing demand planning.

Designs parameterized quantum circuits (ansatz) for variational quantum algorithms, enabling optimization in quantum computing applications.

Analyzes IFC building models for validation, property extraction, and model comparison to ensure data integrity in BIM workflows.

Forecasts healthcare workforce requirements by analyzing patient volume trends, acuity levels, productivity standards, and skill mix optimization.

Provides a Goal-Oriented Action Planning (GOAP) system for AI agents to determine optimal action sequences based on world state and action costs.