Albert Alov

AI Infrastructure Engineer building production LLM pipelines and reliability systems for distributed platforms. Recently focused on AI/LLM systems: evaluation frameworks, structured output validation

Montevideo Joined on Mar 20, 2026 alikalov@gmail.com https://elsapocripto.com/

Pronouns

Albert

Albert Alov

Apr 4

We traced an MCP server calling an LLM — both sides, one trace

#ai #webdev #programming #javascript

7 min read

Want to connect with Albert Alov?

Create an account to connect with Albert Alov. You can also sign in below to proceed if you already have an account.

Create Account

Already have an account? Sign in

Albert Alov

Apr 4

Every PostgreSQL MCP server eats your context window. Here's how I collapsed 4 into 1.

#ai #postgres #mcp #typescript

7 min read

Albert Alov

Mar 30

MCP servers are the fastest-growing part of the AI stack. They have zero observability.

#ai #mcp #opentelemetry #typescript

5 min read

Albert Alov

Mar 29

Your LLM traces are write-only

#ai #programming #typescript #opentelemetry

5 min read

Albert Alov

Mar 27

Your AI agent re-sends 80% of your budget every loop

#ai #opentelemetry #typescript #observability

5 min read

Albert Alov

Mar 24

Your LLM streaming traces are lying to you

#ai #opentelemetry #typescript #observability

6 min read

Albert Alov

Mar 21

OpenTelemetry just standardized LLM tracing. Here's what it actually looks like in code.

#ai #observability #typescript #opentelemetry

6 min read

Albert Alov

Mar 21

I audited my tool, fixed 44 bugs - and it still didn’t work

#ai #webdev #opentelemetry #typescript

5 min read

Albert Alov

Mar 21

My AI bot burned through my API budget overnight. So I built an open-source tool to make sure it never happens again.

#ai #opensource #opentelemetry #typescript

5 min read

DEV Community

Albert Alov

Badges

1 Week Community Wellness Streak

Writing Debut

Skills/Languages

Currently learning

Currently hacking on

Available for

We traced an MCP server calling an LLM — both sides, one trace

Want to connect with Albert Alov?

Every PostgreSQL MCP server eats your context window. Here's how I collapsed 4 into 1.

MCP servers are the fastest-growing part of the AI stack. They have zero observability.

Your LLM traces are write-only

Your AI agent re-sends 80% of your budget every loop

Your LLM streaming traces are lying to you

OpenTelemetry just standardized LLM tracing. Here's what it actually looks like in code.

I audited my tool, fixed 44 bugs - and it still didn’t work

My AI bot burned through my API budget overnight. So I built an open-source tool to make sure it never happens again.