Ricardo Feldhaus Krisanoski

I build AI-assisted operating systems for CRM, WhatsApp sales, revenue operations and internal workflows.

Joined on May 28, 2026

Oct 31

A Public Case Study Should Be an Evidence Map

#ai #machinelearning #architecture #writing

3 min read

Want to connect with Ricardo Feldhaus Krisanoski?

Create an account to connect with Ricardo Feldhaus Krisanoski. You can also sign in below to proceed if you already have an account.

Create Account

Already have an account? Sign in

Ricardo Feldhaus Krisanoski

Oct 30

What a Small RAG Fine-Tuning Experiment Can Actually Prove

#ai #machinelearning #python #rag

6 min read

Ricardo Feldhaus Krisanoski

Oct 29

Your AI Workflow Needs an Exception Contract

#ai #architecture #observability #productivity

3 min read

Ricardo Feldhaus Krisanoski

Oct 28

The Cost of Changing Your Automation Is Part of Its Design

#architecture #ai #productivity #devops

3 min read

Ricardo Feldhaus Krisanoski

Oct 27

A Retrieval Result Needs an Allowed-Use Field

#rag #ai #architecture #security

3 min read

Ricardo Feldhaus Krisanoski

Oct 26

A Read-Only Integration Is a Deliberate Product Boundary

#architecture #api #security #productivity

3 min read

Ricardo Feldhaus Krisanoski

Oct 25

Make Handoffs Explainable Before You Make Them Automatic

#ai #architecture #productivity #ux

3 min read

Ricardo Feldhaus Krisanoski

Oct 24

Freeze the Comparison, Not the Working Document

#architecture #database #productivity #ai

3 min read

Ricardo Feldhaus Krisanoski

Oct 23

What a Production Brief Should Say When the Data Is Incomplete

#architecture #productivity #ai

3 min read

Ricardo Feldhaus Krisanoski

Oct 22

Read Access and Action Authority Belong to Different Contracts

#architecture #ai #programming

3 min read

Ricardo Feldhaus Krisanoski

Oct 21

Do Not Turn an Observability Failure into a Business Outage

#architecture #programming #ai

3 min read

Ricardo Feldhaus Krisanoski

Oct 20

A Retry Needs an Identity, Not Just Another Attempt

#architecture #programming #webdev

3 min read

Ricardo Feldhaus Krisanoski

Oct 19

An Approval Queue Is a Product Interface

#architecture #ai #productivity

3 min read

Ricardo Feldhaus Krisanoski

Oct 18

A Quote Comparison Needs an Equivalence Policy

#architecture #programming #productivity

3 min read

Ricardo Feldhaus Krisanoski

Oct 17

A Useful Context Packet Should Explain What It Left Out

#ai #architecture #programming

3 min read

Ricardo Feldhaus Krisanoski

Oct 16

An Evaluation Score Is Only as Useful as Its Target

#python #machinelearning #datascience #ai

5 min read

Ricardo Feldhaus Krisanoski

Sep 23

A Daily Production Brief Is a Coordination Interface, Not a Dashboard

#ai #architecture #productivity #ecommerce

4 min read

Ricardo Feldhaus Krisanoski

Sep 16

Operational Knowledge Is a Product Surface, Not a Folder

#productivity

4 min read

Ricardo Feldhaus Krisanoski

Aug 16

What I log from AI workflows before I trust the automation

#ai #observability #n8n #architecture

3 min read

Ricardo Feldhaus Krisanoski

Aug 13

AI Coding Agents Need a Decision Gate Before the Implementation Loop

#ai #opensource #architecture #productivity

5 min read

Ricardo Feldhaus Krisanoski

Jun 13

Why my first RAG layer starts in Postgres, not in a standalone vector database

#ai #rag #postgres #architecture

3 min read

Ricardo Feldhaus Krisanoski

May 28

I added a context resolver before an AI sales agent replies

#ai #automation #crm #architecture

3 min read

DEV Community

Ricardo Feldhaus Krisanoski

Badges

1 Week Community Wellness Streak

Writing Debut

A Public Case Study Should Be an Evidence Map

Want to connect with Ricardo Feldhaus Krisanoski?

What a Small RAG Fine-Tuning Experiment Can Actually Prove

Your AI Workflow Needs an Exception Contract

The Cost of Changing Your Automation Is Part of Its Design

A Retrieval Result Needs an Allowed-Use Field

A Read-Only Integration Is a Deliberate Product Boundary

Make Handoffs Explainable Before You Make Them Automatic

Freeze the Comparison, Not the Working Document

What a Production Brief Should Say When the Data Is Incomplete

Read Access and Action Authority Belong to Different Contracts

Do Not Turn an Observability Failure into a Business Outage

A Retry Needs an Identity, Not Just Another Attempt

An Approval Queue Is a Product Interface

A Quote Comparison Needs an Equivalence Policy

A Useful Context Packet Should Explain What It Left Out

An Evaluation Score Is Only as Useful as Its Target

A Daily Production Brief Is a Coordination Interface, Not a Dashboard

Operational Knowledge Is a Product Surface, Not a Folder

What I log from AI workflows before I trust the automation

AI Coding Agents Need a Decision Gate Before the Implementation Loop

Why my first RAG layer starts in Postgres, not in a standalone vector database

I added a context resolver before an AI sales agent replies