RAJANAND ILANGOVAN
@rajanand.org
Lead Data Engineer ==> Data Architect
https://rajanand.org/substack
https://rajanand.org/substack
This is what strategy looks like in engineering:
choosing failure modes before writing code.
If you want more influence, stop describing pipelines as code.
Start explaining which risks you removed, and which ones you accepted.
🚀 Read my newsletter rajanand.substack.com/welcome
choosing failure modes before writing code.
If you want more influence, stop describing pipelines as code.
Start explaining which risks you removed, and which ones you accepted.
🚀 Read my newsletter rajanand.substack.com/welcome
January 1, 2026 at 7:40 PM
This is what strategy looks like in engineering:
choosing failure modes before writing code.
If you want more influence, stop describing pipelines as code.
Start explaining which risks you removed, and which ones you accepted.
🚀 Read my newsletter rajanand.substack.com/welcome
choosing failure modes before writing code.
If you want more influence, stop describing pipelines as code.
Start explaining which risks you removed, and which ones you accepted.
🚀 Read my newsletter rajanand.substack.com/welcome
Example: a third-party API.
You can bet the schema won’t change, or design as if it will.
Immutability, idempotency, contracts aren’t "best practices."
They’re risk controls.
They decide whether failure is cheap or catastrophic.
You can bet the schema won’t change, or design as if it will.
Immutability, idempotency, contracts aren’t "best practices."
They’re risk controls.
They decide whether failure is cheap or catastrophic.
January 1, 2026 at 7:40 PM
Example: a third-party API.
You can bet the schema won’t change, or design as if it will.
Immutability, idempotency, contracts aren’t "best practices."
They’re risk controls.
They decide whether failure is cheap or catastrophic.
You can bet the schema won’t change, or design as if it will.
Immutability, idempotency, contracts aren’t "best practices."
They’re risk controls.
They decide whether failure is cheap or catastrophic.
Senior engineers ask a different question:
"What’s the cost of being wrong if this assumption fails?"
"What’s the cost of being wrong if this assumption fails?"
January 1, 2026 at 7:40 PM
Senior engineers ask a different question:
"What’s the cost of being wrong if this assumption fails?"
"What’s the cost of being wrong if this assumption fails?"
If you want your ideas to land, change the framing:
Don’t argue why it’s better.
Explain which constraint it relieves.
That’s how senior thinkers decide.
🚀 Read my newsletter here bit.ly/49i2qTc
Don’t argue why it’s better.
Explain which constraint it relieves.
That’s how senior thinkers decide.
🚀 Read my newsletter here bit.ly/49i2qTc
January 1, 2026 at 2:56 PM
If you want your ideas to land, change the framing:
Don’t argue why it’s better.
Explain which constraint it relieves.
That’s how senior thinkers decide.
🚀 Read my newsletter here bit.ly/49i2qTc
Don’t argue why it’s better.
Explain which constraint it relieves.
That’s how senior thinkers decide.
🚀 Read my newsletter here bit.ly/49i2qTc
The hidden constraint:
“We can’t afford disruption this quarter.”
Revenue targets, customer commitments, and org focus
make stability more valuable than elegance.
“We can’t afford disruption this quarter.”
Revenue targets, customer commitments, and org focus
make stability more valuable than elegance.
January 1, 2026 at 2:56 PM
The hidden constraint:
“We can’t afford disruption this quarter.”
Revenue targets, customer commitments, and org focus
make stability more valuable than elegance.
“We can’t afford disruption this quarter.”
Revenue targets, customer commitments, and org focus
make stability more valuable than elegance.
Example:
You propose a clean re-architecture.
It’s cheaper long-term and clearly better design.
Leadership says no.
Not because they don’t see the value.
You propose a clean re-architecture.
It’s cheaper long-term and clearly better design.
Leadership says no.
Not because they don’t see the value.
January 1, 2026 at 2:56 PM
Example:
You propose a clean re-architecture.
It’s cheaper long-term and clearly better design.
Leadership says no.
Not because they don’t see the value.
You propose a clean re-architecture.
It’s cheaper long-term and clearly better design.
Leadership says no.
Not because they don’t see the value.
The real constraints leaders optimize for:
time, risk, attention, cost, and reversibility.
Your idea can be technically correct
and still be the wrong decision right now.
time, risk, attention, cost, and reversibility.
Your idea can be technically correct
and still be the wrong decision right now.
January 1, 2026 at 2:56 PM
The real constraints leaders optimize for:
time, risk, attention, cost, and reversibility.
Your idea can be technically correct
and still be the wrong decision right now.
time, risk, attention, cost, and reversibility.
Your idea can be technically correct
and still be the wrong decision right now.
This is what strategy looks like in engineering:
choosing failure modes before writing code.
If you want more influence, stop describing pipelines as code.
Start explaining which risks you removed, and which ones you accepted.
🚀 Read my newsletter rajanand.substack.com/welcome
choosing failure modes before writing code.
If you want more influence, stop describing pipelines as code.
Start explaining which risks you removed, and which ones you accepted.
🚀 Read my newsletter rajanand.substack.com/welcome
January 1, 2026 at 1:30 PM
This is what strategy looks like in engineering:
choosing failure modes before writing code.
If you want more influence, stop describing pipelines as code.
Start explaining which risks you removed, and which ones you accepted.
🚀 Read my newsletter rajanand.substack.com/welcome
choosing failure modes before writing code.
If you want more influence, stop describing pipelines as code.
Start explaining which risks you removed, and which ones you accepted.
🚀 Read my newsletter rajanand.substack.com/welcome
Example: a third-party API.
You can bet the schema won’t change, or design as if it will.
Immutability, idempotency, contracts aren’t "best practices."
They’re risk controls.
They decide whether failure is cheap or catastrophic.
You can bet the schema won’t change, or design as if it will.
Immutability, idempotency, contracts aren’t "best practices."
They’re risk controls.
They decide whether failure is cheap or catastrophic.
January 1, 2026 at 1:30 PM
Example: a third-party API.
You can bet the schema won’t change, or design as if it will.
Immutability, idempotency, contracts aren’t "best practices."
They’re risk controls.
They decide whether failure is cheap or catastrophic.
You can bet the schema won’t change, or design as if it will.
Immutability, idempotency, contracts aren’t "best practices."
They’re risk controls.
They decide whether failure is cheap or catastrophic.
Senior engineers ask a different question:
"What’s the cost of being wrong if this assumption fails?"
"What’s the cost of being wrong if this assumption fails?"
January 1, 2026 at 1:30 PM
Senior engineers ask a different question:
"What’s the cost of being wrong if this assumption fails?"
"What’s the cost of being wrong if this assumption fails?"
No, 2026 is the current year, not next year.
Happy new year!
Happy new year!
January 1, 2026 at 9:13 AM
No, 2026 is the current year, not next year.
Happy new year!
Happy new year!
This resonates with me.
Writing when I can’t decide, and walking when my thinking is cloudy, are my default resets. They work far more often than trying harder.
Writing when I can’t decide, and walking when my thinking is cloudy, are my default resets. They work far more often than trying harder.
January 1, 2026 at 9:07 AM
This resonates with me.
Writing when I can’t decide, and walking when my thinking is cloudy, are my default resets. They work far more often than trying harder.
Writing when I can’t decide, and walking when my thinking is cloudy, are my default resets. They work far more often than trying harder.