3. Data, Science & AI

Deploys AI models using NVIDIA NIM with OpenAI-compatible APIs for self-hosted or cloud environments.

Automates review of database schema, queries, and migrations to enforce normalization, indexing, and optimization best practices.

Provides essential PostgreSQL extensions for data analysis, text search, and geospatial processing.

Builds autonomous AI agents via Claude Agent SDK with tool calling, MCP integration, and production best practices.

Provides patterns and best practices for coordinating multiple AI agents, decomposing tasks, and orchestrating workflows in agent-based systems.

Performs unsupervised learning tasks including clustering, dimensionality reduction, and anomaly detection to uncover hidden patterns in unlabeled datasets.

Enables AI models to process and generate responses across vision, audio, and document inputs for comprehensive multimodal understanding.

Builds and trains deep learning models including MLPs and CNNs using PyTorch with industry best practices.

Expertise in building enterprise AI applications using Azure OpenAI, Cognitive Services, and Azure Machine Learning with scalable architectural patterns.

Automates the updating process for Large Language Models to ensure access to the latest model versions and features.

Architects and deploys AI workloads across AWS, Azure, GCP, and OCI with intelligent routing and cost optimization.

Builds end-to-end AI applications from data pipelines to production using LLMs, agents, and modern frameworks.

Builds AI solutions for image classification, object detection, and transfer learning in computer vision applications.

Expert in designing and optimizing Retrieval-Augmented Generation systems for efficient knowledge retrieval and generation in AI applications.

Trains and fine-tunes LLMs using HuggingFace Transformers, TRL, and cloud GPUs with SFT, DPO, and GRPO methods.

Builds production-ready supervised learning models for classification and regression with hyperparameter optimization.

Provides foundational knowledge and practical guidance on machine learning algorithms, data preprocessing, feature engineering, and model evaluation techniques.

Provides production-grade AI architecture patterns for enterprise, emphasizing security, governance, scalability, and operational excellence.

Sets up Linux environments with TensorFlow, PyTorch, CUDA, and Jupyter for data science and AI/ML development.

Enables building intelligent systems through machine learning, data engineering, LLMs, and MLOps using Python.

Provides text processing and analysis using modern NLP techniques including preprocessing, embeddings, and transformers.

Intelligently analyzes and synchronizes data between overview listings and actual group headers using agent-assisted processing for accurate data alignment.

Transforms and exports data using DuckDB SQL, supporting multiple formats and complex operations like joins, aggregations, and format conversion for efficient data processing.

Designs scalable data architectures with modeling, pipeline construction, and governance policies for efficient data management.