Hi everyone,

I’m preparing for interviews and one question that keeps coming up is:

“How would you migrate from Informatica (or ADF / Databricks) to Microsoft Fabric?”

Our environment is mostly:

• On-prem SQL Server as source
• Some files (CSV / Excel / shared drive)
• Existing ETL built in Informatica / ADF / Databricks
• Looking at Microsoft Fabric as target

I’m trying to understand how teams approach this in real projects.

My confusion is:

1. Data migration → moving source data/files into Fabric (Lakehouse / Warehouse)
2. Tool migration → rebuilding ETL pipelines / workflows / transformations from Informatica / ADF / notebooks into Fabric

Questions:

• What’s your migration approach?
• Do you move data first or rebuild pipelines first?
• For on-prem SQL Server, are you using gateway / mirroring / pipelines?
• How much can be automated vs manually rebuilt?
• Has anyone used migration accelerators like Kanerika for Informatica → Fabric? What exactly do they automate?

Would really appreciate practical examples or interview-ready answers.

Thanks!

Hi everyone,

I’m preparing for interviews and one question that keeps coming up is:

“How would you migrate from Informatica (or ADF / Databricks) to Microsoft Fabric?”

Our environment is mostly:

• On-prem SQL Server as source
• Some files (CSV / Excel / shared drive)
• Existing ETL built in Informatica / ADF / Databricks
• Looking at Microsoft Fabric as target

I’m trying to understand how teams approach this in real projects.

My confusion is:

1. Data migration → moving source data/files into Fabric (Lakehouse / Warehouse)
2. Tool migration → rebuilding ETL pipelines / workflows / transformations from Informatica / ADF / notebooks into Fabric

Questions:

• What’s your migration approach?
• Do you move data first or rebuild pipelines first?
• For on-prem SQL Server, are you using gateway / mirroring / pipelines?
• How much can be automated vs manually rebuilt?
• Has anyone used migration accelerators like Kanerika for Informatica → Fabric? What exactly do they automate?

Would really appreciate practical examples or interview-ready answers.

Thanks!

Hi everyone,

I had an interview question around orchestrating Databricks notebooks from Azure Data Factory and wanted to hear real-world approaches.

Scenario:

We have 50+ Databricks notebooks across Bronze / Silver / Gold layers.

Example:

/bronze/customer
/bronze/orders
/silver/customer
/silver/orders
/gold/sales_kpi
...

ADF is the main orchestrator.

My first thought was:

• add one Databricks Notebook activity per notebook

…but with 50+ notebooks that feels messy very quickly:

• huge ADF pipeline UI
• hard to maintain dependencies
• painful when onboarding a new notebook
• duplicate parameters/config

How do teams usually solve this in production?

Questions:

1. Do you call one master notebook from ADF and then use dbutils.notebook.run() inside Databricks??

If yes,, how we do it? in databricks and adf

and also lets suppose we are not using databricks workflow for now...

1. How do you handle:
   • notebook dependencies
   • retries
   • parallel vs sequential execution
   • passing parameters like run_date / batch_id
2. At what point do you stop orchestrating inside ADF and move orchestration fully into Databricks?

Would love to hear what’s working in real production environments, especially in Azure.

I had originally planned to attend my US visitor visa interview in Feb 2026, but due to the recent visa bond discussions/rules, I skipped it at that time.

Now I’m seeing updates that World Cup ticket holders may not require the bond requirement in certain cases.

I’m trying to understand:

1. If we already have FIFA World Cup tickets, how fast can we realistically get a US visa interview appointment from Canada?
2. Do we still need to go through the full B1/B2 process/interview again as normal?
3. Are Canada PR holders getting any faster processing or special consideration for World Cup travel?

Would appreciate hearing from anyone applying from Toronto/Vancouver/Montreal recently.