Bobur Umurzokov

1 week ago | dev.to | Bobur Umurzokov

When you're building a startup, you move fast. You want to test ideas quickly, keep costs low, and be ready to change directions anytime. That’s why the tech stack you choose early on matters a lot. What works for big companies often doesn’t fit early-stage startups. If your product is still new and you're exploring the market, don’t build a complicated architecture too soon. You don’t need microservices for everything. Instead, start with one backend and one frontend that handles everything.

2 weeks ago | dev.to | Bobur Umurzokov

AI agents are getting a lot of attention lately, but getting started can be confusing. You may have heard about tools like LangChain/LangGraph, Semantic Kernel, AutoGen, or LlamaIndex. They are powerful, but sometimes all you need is something simple that works. Azure AI Agent Service lets you build AI agents that can use your own tools, such as a function to read data from a Postgres database.

2 months ago | medium.com | Bobur Umurzokov

Bobur·Follow7 min read·--Versioned APIs with Dynamic DatabasesDuring new feature development, we frequently introduce new API versions to roll out new features gradually while keeping the existing API functional for the connected client applications. Oftentimes, the new API version has a new database schema change and we need to serve different datasets based on the API version.

2 months ago | dzone.com | Bobur Umurzokov

Neon is now available on the Azure marketplace. The new integration between Neon and Azure allows you to manage your Neon subscription and billing through the Azure portal as if Neon were an Azure product. Azure serverless and Neon are a natural combination — Azure serverless frees you from managing your web server infrastructure. Neon does the same for databases, offering additional features like data branching and vector database extensions.

2 months ago | dev.to | Import Os |Bobur Umurzokov

During new feature development, we frequently introduce new API versions to roll out new features gradually while keeping the existing API functional for the connected client applications. Oftentimes, the new API version has a new database schema change and we need to serve different datasets based on the API version. In such scenarios, we can use Neon’s branching feature to dynamically manage database versions instead of creating separate databases or handling costly data migrations.