Google

1. Single-Threaded Leader for FastCloud (FCCY) with CEO visibility, led end-to-end delivery across 25+ global teams, owning delivery of 2 services directly. Reduce latency by 85% and realised $1B+ revenue opportunities for all datacenter.

2. Led Change Management across reorgs and drove alignment at VP level, and maintained attrition rate to less than 10% as compared to 30% attrition in peer teams, and delivered multi-year FCCY program on schedule.

3. Established Capacity Planning process, conceptualized and built order-validation tooling, mitigating 40% of recurring delivery risks and improving p95 SLO coverage from 58% to 100% since Nov 2024.

4. Architected and delivered GDPR compliance Data Center TurnUp for TPC Cloud-level OKR, enabling $100B+ European market access (TAM). Improving reliability from ~90% to 99.9%, shifting to plug-and-play architecture, re-using existing components.

5. Conducted 150+ interviews in 2.5 years. Led 14+ Hiring Committees (HCs) at Google, made unbiased hire/no hire decision based on interview feedback.

6. Developed leadership talent by mentoring 30+ engineers (L3/L4), coached 3 first-time manager.

7. Awarded Outstanding rating in 2024 (top 18%), rated Excellent Manager (4.95/5) across all dimensions: Deliver Results, Develop People, Build Community. Enabled 5 promotions over 3 years.

Engineering Manager - AI and Computing Infrastructure

Flipkart

During my 5-month tenure as a leader of the conversational chat assistant platform, I achieved the following:

1. Unified 13 engineers from two teams into a high-performing Conversational Assistant unit, doubling sprint throughput in the first quarter, and accelerated project timelines.

2. Increased product adoption by 500% through 13 targeted A/B tests, and developed the 2022 strategic roadmap to achieve 10x growth in user adoption.

Tech Lead - User Acquisition and Retention

Sumyag Data Sciences

1. Founded and scaled an AI solutions company, managing 35 active client accounts and generating $1M annual revenue with $198K profit in 2020.

2. Redesigned core architecture for distributed computing and parallel processing, cutting onboarding from 12 to 3 weeks and adding 20+ clients. Created product roadmap for ExdionPOD, growing customer base from 4 to 15.

3. Hired and led a 22-member cross-functional team across engineering, data science, and product, overseeing delivery from ideation to launch. Named Top 16 InsurTech in USA by the Council of Insurance Agents and Brokers (CIAB).

Partner - Co Founder

@WalmartLabs India

1. Established the Collibra development team in India by collaborating with the Director of Data Governance on a 12-month product roadmap.

2. Recruited and managed a 15-member global team to deliver REST services integrating Collibra with internal systems, improving data accuracy from 67% to 95% and achieving a 5/5 customer satisfaction score in all deployments.

Senior Software Engineer - Data Governance

1. Ranked 6th in global Collibra Experts leaderboard 2015-16 (ranked above Collibra product developers).

2. Generated Business Glossary and Issue Management system in Collibra, improving Data Governance policy adherence by 70%.

Associate Programmer Analyst - Data Quality & Governance

Hands-on technical engineering manager with 11+ years experience, including 6+ years as a people manager.

At Google, I’ve driven $1B+ in revenue opportunities through large-scale cloud initiatives, led teams of 25+ engineers, and managed programs spanning 25+ global teams. Consistently rated among the highest-performing managers, with recognition for leadership, innovation, and cross-org delivery and earning top-tier annual and managerial performance ratings at Google.

Before Google, I founded and scaled a profitable AI solutions company to $1M ARR and was named Top 16 InsurTech in the USA by CIAB. 
My leadership style blends design thinking, system design expertise, and a bias for building high-impact teams and products from the ground up.

Core strengths:
1. Strategic Skills: Change Management, Strategic Roadmapping, Program Governance, Risk Management, Process Improvements, Stakeholder Management, OKRs.

2. Org Building Skills: Hiring, Building and Managing Teams, Retention, Career Growth, Coaching, Conflict Resolution, Inclusion, Cross-functional Alignment, Design Thinking, Organisational Development, Change Management.

3. Technical: System Design, AWS, PostgreSQL, A/B Testing, Software Design, Distributed Systems, Python, Microservices Architecture, CI/CD, Cloud Architecture, SLO Management, High Availability, NLP, AI/ML.

4. Product Skills: Product Roadmaps, Release Management, Customer Interactions, JIRA, Requirements Gathering, Agile.

Outside of work: I’m a Top 30 writer in Management & Leadership in India (158K+ followers), and ranked #1 content creator in my niche (Mar 2025) as per Favikon rankings.

Give me 2 minutes and I'll teach you how Netflix never crashes.

100M people binge-watch the same show. 

And yet… Netflix never blinks. No downtime. No lag.

Meanwhile, your codebase breaks under a few thousand users.


Here’s what Netflix gets right (that we should all steal):

1. Microservices 

Each component (login, payments, recommendations) is its own service.

No single point of failure. No domino effect.

2. Open Connect CDN

They physically ship servers worldwide. 

Content loads from your nearest location. Buffering? What buffering?

3. Chaos Engineering 

Netflix built Chaos Monkey — a tool that kills their own servers at random.

If it survives monkeys, it survives Monday traffic.

4. Personalization at scale 

Your “Top Picks” are powered by real-time ML over Kafka, Spark, Cassandra.

Every click feeds the system. Every second is personalized.

5. Observability 

Grafana. ELK. Real-time alerts.

They don’t guess what’s broken. They know before it happens.


This isn’t just good engineering.

It’s resilience, built brick-by-brick.

If you're building for millions, study Netflix.

If you're building for ten users, still study Netflix.



♻️ Save this for your next system design round

🔔 Follow Abhishek Kumar for more tech breakdowns without the buzzwords.

Abhishek Kumar

They promoted someone who wrote worse code than me.
Stanford taught me why.
Here's the game most engineers don't see:

I had better technical output.

They had better business communication.

The shift:

Engineers stuck at mid-level think: "My code speaks for itself."

Engineers who get promoted think: "Let me translate this for stakeholders."

Stanford's framework:

Before any project, answer:

- What breaks without this? (Business problem, not technical challenge)
- Whose goals does this move? (Find stakeholders, not just your manager)
- What's the ₹ impact? (Money saved or earned, not just performance metrics)
- Who needs to know? (Visibility is part of the job)

Example:

Before: "Reduced latency by 200ms"
After: "Fixed checkout timeouts costing ₹8Cr in abandoned carts"

Same work. Different language.

One gets you "good job." One gets you promoted.

The truth:

Your manager promotes people who speak business, not just code.

Revenue. Risk. Results.

Learn to translate. Or stay stuck.

Are you optimizing for code quality or business clarity?

Follow Abhishek Kumar for real-world problem specific posts !!

I have been broke many times in my life.

Today, it all seems like an overnight success.

But the truth is -

2010: I couldn't clear IIT.

2014: My first job was comparing Excel sheets.

2017: Bought my first car - Honda City and house 

2021: My first start-up shut down.

2020:I had a debt of 28 Lakhs 

2022: I Joined Google and my life turned around for the better. 

2022: Invested money and bought my dream Car - Mercedes

2024: Mentored 1000+ students and spoke at events helping students do the same.

2024: Took my parents on their first international trip !!

People only know the happy days.

The struggle to survive each day behind that is the actual part of life. 

The ups and downs of life have taught me more than ever.

What helped me came out of that rut:

▶️Perseverance
▶️ Self - believe 
▶️Determination

If I can do it, You can too.

What is your biggest challenge?

I can tell if someone's from a service company within 5 minutes of an interview.

It's not their resume. It's how they freeze.

I've interviewed 50+ engineers making the service-to-product jump.
The pattern is consistent:

Me: "Design a URL shortener."
Them: long pause "Can you give me more details on the requirements?"

The problem isn't technical skill. 

It's that they've been trained to execute, not to think.

Service company workflow: Get JIRA ticket → Read complete spec → Code → Submit for review

Product interview: Here's a vague problem → Make assumptions → Make trade-offs → Justify your choices → Build

Your brain is wired for one. The interview tests the other.

Here's the pattern I see:

1. You wait for complete clarity before proceeding

Interview expects: "Based on typical URL shortener requirements, I'll assume X, Y, Z. Let me know if that changes."

You should keep asking clarifying questions instead of making progress.

2. You give one solution, no alternatives

Interview: "How would you implement rate limiting?"

Reactive answer: "Use a counter in Redis and check if it exceeds the threshold."

Proactive answer: "Three approaches: Token bucket allows traffic bursts, leaky bucket smooths requests, fixed window is simplest to implement. 

For an API that needs to handle sudden spikes, I'd use a token bucket. Here's the trade-off..."

3. You can't justify why you chose something

Interview: "Why SQL over NoSQL here?"

Reactive: "Because the data is structured."

Proactive: "This use case has complex joins for analytics and needs ACID guarantees for payments. NoSQL would give us horizontal scaling, but we don't need that at this scale. 

SQL's transactional consistency matters more. If we hit 10M+ users, we revisit."

How to reactivate your thinking muscle:

Exercise 1: 

Every work task, ask: "Why this way? What's an alternative? What are trade-offs?"

Even if you're following a spec, reverse-engineer the decisions.

Exercise 2: 
Design simple systems (URL shortener, notification system) with no spec. 

Force yourself to make assumptions. 
Defend them out loud. 
Record yourself. 
Notice where you go vague.

Exercise 3: 

Practice articulating trade-offs for common decisions

- SQL vs NoSQL
- REST vs GraphQL
- Sync vs Async processing
- Caching strategies

Say them out loud. Not in your head.


Timeline: 8 weeks of deliberate practice.

One engineer I mentored failed 3 product interviews.

He spent 8 weeks doing these exercises every single day.

Interview 4: Offer at 22 LPA from a Bangalore fintech.


What's one technical decision you made this week that you can't explain the trade-offs for?

Follow Abhishek Kumar for systems thinking and interview strategies.

He celebrated his ₹25L offer for 2 weeks.
Then discovered his junior got ₹35L.

Same company. Same role. Same week.

An engineer I know switched to a Bangalore unicorn last year.

Asked for: ₹22L
Got: ₹25L

Felt like he won.

Posted on LinkedIn. Called his parents. Finally made it.

Two weeks later, his junior with 1 year less experience joined the same company.

Over coffee, it came up.

Junior's offer: ₹35L.

Same role. Same level. Same bar.

Only difference?

He asked for ₹22L based on "what felt reasonable."

Junior asked for ₹35L based on market research.

That ₹10L gap isn't one-time.

It compounds with every raise. Every future switch.

Over 5 years? That's ₹40-50L left on the table.

This isn't a service company problem.

I've seen Google engineers accept the first offer without countering.

Amazon L5s who don't know competing offers exist.

Senior engineers at unicorns who anchor on their last CTC.

The pattern is the same:

"What feels reasonable" ≠ Market value

Product companies have salary bands.

They have budget allocated.

They're ready to pay market rate.

But if you anchor low, they'll meet you there.

Why would they correct upward?

Here's what his junior did differently (swipe) ↓

It's not luck. It's knowing the game.


Follow Abhishek Kumar for frameworks that close the information gap between you and companies.

Salary Negotiation

Last year, 12 engineers from my network quit to prep for product companies.

1 got into Amazon at ₹42L. 11 ran out of savings and went back to their old jobs.

Same education. Similar skills. All equally desperate to escape.

The difference wasn't talent.
It was the structure.

The one who made it treated the break like a 12-week sprint. 
The 11 who failed treated it like time off.

Here's what I noticed:

Week 1: Everyone's motivated. 
Week 4: Half are sleeping till 11 AM. 
Week 8: Most are panicking about money. 
Week 12: Only one is still following a system.

I've mentored 40+ engineers through career breaks now.

The pattern is clear: Quitting works when you have a framework. 

Quitting fails when you're running on hope and frustration.

If you're stuck watching juniors get ₹25L offers while you're at ₹14L.
And quitting feels like the only option.

Don't do it out of emotion.
Do it with a plan.

The Strategic Break Framework 👇


(Swipe to see the 6-step system that separates the 1 who makes it from the 11 who don't)


Follow Abhishek Kumar for strategic engineering career transitions.

Prep Framework

I stayed 2.5 years at my first company before I realized it was a dead-end.

The signals were there at Month 6. I just didn't know what to look for.

Not all first jobs are equal.

Some are "learn fast, leave faster": great for 18 months, terrible long-term.
Some are "grow and stay": real paths, mentorship, promotions.

You can tell by Month 6.

Here are the signals to check:

1. Growth path signals:

- Ask: "What does the next level look like?" Vague answer? Red flag.
- Are seniors promoted internally or hired externally?
- Still learning or just faster at same tasks?

2. Work quality signals: 

- What do 2-3 year engineers do? Still fixing bugs or owning features? 
- Code reviews: Actual feedback or just "LGTM" approvals?

3. Investment signals: 

- Does the learning budget exist and gets used?
- How's attrition handled? Professional or betrayal?
- 1-on-1s: Growth conversations or rushed check-ins?

4. Framework signals:

- 5+ red flags → Learn fast, leave by Month 18-24
- Mixed signals → Re-evaluate at Month 12 
- 5+ green flags → Stay 3-4 years, aim for promotions


I ignored red flags at my first company.

"It'll get better" I told myself.
It didn't.

Left at Year 2.5 when I should've left at Year 1.5.

Cost me 12-15 months of better learning and ₹3-4L.

Don't make my mistake.

Evaluate by Month 6. Decide by Month 12. Act by Month 18.

What signals are you seeing at your company?

Follow Abhishek Kumar for more structured growth systems for Indian engineers.

Excellent performance reviews at service company.
Rejected by product company in 1st round.
Here's why:

A junior I mentored spent 18 months at a Bangalore service company writing Java.

Tickets closed on time.
Manager gave "exceeds expectations."
He felt ready.

Then he interviewed at a product company.

Interviewer: "Walk me through your testing strategy."
Him: "QA handles that."
Interviewer: "You've never written a unit test?"

He hadn't. Didn't know he was supposed to.
Interview over in 30 minutes.
This isn't rare.

If you've only worked in one environment, you assume it's normal. 
- Manual deployments. 
- Checkbox code reviews. 
- "Testing" means clicking buttons.

Performance reviews don't catch these gaps. Interviews do.

So here's what to do before your next interview exposes what you don't know:

Ask 3 questions:

1. "What does 'good code' look like?"

Pick any open-source repo. Read 5 merged PRs. 
Notice what reviewers care about that you never considered.

2. "What practices have I never touched?"

Automated tests? CI/CD? Monitoring? List them. 
That's your gap. Pick ONE and add it to a side project this weekend.

3. "What do engineers ahead of me know?"

DM 2 people on LinkedIn: "I'm at [company] with [X] experience. What do you wish you'd learned earlier?"

That junior? Fixed his gaps in 6 months. Got into a product company with a 70% jump.

The difference between 8 LPA and 20+ LPA isn't talent. 
It's asking the right questions before interviews expose what you don't know.

Which question are you avoiding?

Follow Abhishek Kumar for real career frameworks, not fluff.

Indian government: We will build highways to reduce travel time by X hours 

Le truck drivers: Hold my Lassi !!

They knew the answer. 
Still got rejected. 
Here's why.

A service company engineer I know prepared for 2 months. 

Practiced system design. Nailed it in mocks.

Then came the interview at a product company:

Interviewer: "How would you design a rate limiter?"

He knows this. Practiced it 10 times.
But he freezes.
Stammers.

Starts explaining, backtracks, over-explains basics, voice getting shakier with each sentence.

"Is this enough? Should I add more? Did I miss something?"

The interviewer's face says it all.
Rejected.

This drives me insane.

I've watched this pattern destroy careers for 3 years. 

Smart engineers who know their stuff but are stuck at 12 LPA because they can't deliver under pressure.

After 5 rejections, that voice in your head becomes permanent: 

"Maybe I don't actually know this."

You do. The problem isn't knowledge. 
It's performance anxiety dressed up as imposter syndrome.

Here's what actually fixes it:

1. Record yourself.

Solve one problem aloud. Play it back. 

You'll hear exactly where you lose confidence—the hesitation, the over-explaining, the "umms."

Now you know what to fix. Verbal clarity is trainable.

2. Use strategic pauses.

When you need a moment: "Let me think through this."

Two seconds of silence beats two minutes of nervous rambling. 

Silence = thinking. Rambling = panic.

3. Practice recovery.

You will stumble. Everyone does.

Say: "Let me restart that" or "Different approach..."

Winners recover. Losers spiral.

4. Apply everywhere.

15-20 companies. 3-4 interviews weekly.

By interview 8, anxiety drops. 

Questions repeat. You're polished. The only way to build interview confidence is volume.

5. Rate yourself after each interview:

Did I explain clearly?
Did I recover from mistakes?
Did I sound confident?

Not "did I get the answer right." That's secondary.

Progress tracking rebuilds confidence faster than perfection ever will.


The gap between you and a product company offer isn't knowledge.
It's delivery.

And delivery is fixable in 3 weeks if you actually practice it.

Stop blaming the algorithm. Start recording yourself.

Which one are you fixing first: the freeze, the rambling, or the recovery?

Follow Abhishek Kumar for frameworks that fix the gaps service companies don't teach.

taplio

Abhishek Kumar's Linkedin Analytics

Abhishek Kumar

Performance ScoreAbhishek Kumar's LinkedIn Performance

What is Abhishek talking about?

Abhishek Kumar's Best Posts (last 30 days)

Ready to grow your LinkedIn presence?

taplio

.css-6ytdur{background-image:linear-gradient(284deg, rgba(39, 43, 72, 1) 0%, rgba(33, 47, 183, 1) 100%);color:transparent;-webkit-background-clip:text;background-clip:text;}Abhishek Kumar's Linkedin Analytics

Abhishek Kumar

Performance Score.css-goub6s{font-size:var(--chakra-fontSizes-md);font-weight:500;color:var(--chakra-colors-gray-600);display:block;margin-top:var(--chakra-space-1);}Abhishek Kumar's LinkedIn Performance

What is Abhishek talking about?

Abhishek Kumar's Best Posts .css-1emoo84{font-size:var(--chakra-fontSizes-md);font-weight:400;color:var(--chakra-colors-gray-600);}(last 30 days)

Ready to grow your LinkedIn presence?

Abhishek Kumar's Linkedin Analytics

Performance ScoreAbhishek Kumar's LinkedIn Performance

Abhishek Kumar's Best Posts (last 30 days)