Saurav Kumar's contributions

Learn how the MCP Toolbox Java SDK supports type-safe AI agent tools for controlled database access, validation, authorization, and business-level operations.

Learn how Google Cloud Spanner can serve as a transactional runtime catalog for Apache Iceberg tables while data and metadata files remain in object storage.

Learn how the MCP Toolbox for Databases Java SDK provides type-safe, business-level database tools for AI agents through the Model Context Protocol.

Learn how Google Cloud Spanner Queues combine database changes and message delivery in a transaction, reducing outbox-pattern complexity for asynchronous microservice workflows.

Learn how Google Cloud Storage uses CRC32C and MD5 checksums to validate uploads and downloads and detect data corruption during transfers.

Learn when to use Google ADK graph workflows instead of a single AI agent, including routing, sequential and parallel execution, joins, deterministic rules, and human review.

Learn when Google Cloud Spanner Omni is appropriate for running a strongly consistent distributed database across AWS, Azure, and other infrastructure environments.

Learn how GitHub Custom Properties add structured business metadata to repositories for ownership, governance, discovery, reporting, and automation.

AI features can be easy to add and surprisingly difficult to debug. A normal API call is fairly simple to follow. A request enters the application, a function runs, and a response comes back. With an AI feature, one user request can involve several model calls, retrieval steps, prompts, tool calls,

Learn how GitHub Custom Properties add structured business metadata to repositories for ownership, criticality, compliance, reporting, governance, and automation.

Suppose your company already has an internal support system. Employees use it to check tickets, look up customer information, and create requests. Now they want to do the same work from Microsoft 365 Copilot. You do not necessarily need to build another chatbot from scratch. Microsoft 365 Copilot ca

Learn how to design AI agent memory across sessions using persistent context, semantic retrieval, memory consolidation, expiration, privacy, and user controls.

AI coding tools have changed quite a bit. They are no longer limited to answering questions or suggesting a line of code. An agent can look at a task, decide what it needs to do, call tools, inspect files, and continue working based on the result. That becomes interesting when you want to put this k

Learn how Chrome 155 changes enterprise policy enforcement for chrome.debugger and how extension developers can handle blocked debugger access and managed browsers.

Learn how OpenAI Prompt Caching works, how to increase cache hits, structure reusable prompt prefixes, monitor cached tokens, and reduce AI API cost.

Learn when to use structured forms with Power Apps and Power Automate, including validation, approvals, data modeling, workflows, and practical architecture decisions.

GitHub is removing legacy SSH algorithms and raising RSA key requirements. Learn which algorithms to check, how to test clients, and what developers need to change.