2016A career in progressPresent

Journey

From solving problems.
To building with responsibility.

Effortless hard workThe work feels natural when the problem is worth solving.

Choose harder challengesEvery difficult problem expands what I can build and who I can help.

10 yearscollege to team leadership

A goal became a passion

Jul 2016 — May 2020 · Lovely Professional University · Jalandhar

4 years

EducationB.Tech in Computer Science · Machine Learning specialization

A 10 LPA ambition started the journey.Curiosity, effortless hard work, and a constant search for harder challenges kept moving the goalpost forward.

Competitive programming, projects, and a Machine Learning specialization turned preparation into passion. The effort was real, but choosing technical challenges I genuinely enjoyed made the hard work feel natural.

MilestonesSelected outcomes
FirstLPU student selected directly for a full-time Amazon role during intern hiring
3 of 3Team Amigos members who received Amazon offers
32 LPAPlacement featured in LPU advertisements across the country
College story

Ten moments that kept moving the goalpost.

Select a moment to follow the ambition, choices, breakthroughs, setbacks, and gratitude that shaped these four years.

Starting point01 / 10

A simple ambition: 10 LPA

I entered college with one clear ambition: earn a 10 LPA offer after graduation. It gave me a starting point, even though I had no idea how far the goal would eventually move.

What keeps me mentoring

The motivation behind every mentoring conversation

Achievements create proud moments, but helping someone discover their potential creates lasting meaning. Krishna’s post reminds me that being present, showing the right direction, and believing in someone can influence an entire journey. That is what keeps me mentoring.

Krishna Barnwal’s LinkedIn post describing Sanjay Gandhi as a down-to-earth mentor who showed him the right direction, recognised his potential, and helped him throughout his placement journey.Open the original post
Explore education details
Full details

B.Tech in Computer Science Engineering — Machine Learning

University
Lovely Professional University, Jalandhar, Punjab
Period
Jul 2016 – May 2020
CGPA
9.06/10
Technologies

Machine LearningData StructuresAlgorithmsCompetitive Programming

Key skills

Problem solvingTeachingMentoring

Explore decisions and interview outcomes
Declined

Campus startup

The offer rose from 8 to 28 LPA. I declined because university rules could close later campus opportunities.

Selected

InterviewBit · Scaler internship

I joined with two Team Amigos friends and helped hire teachers, shape the course, and build the early learning platform.

Selected

Amazon · Direct full-time SDE

Amazon came to LPU for interns. I became the first student selected directly for a full-time role, advertised by the university as 32 LPA.

Rejected

Google · First interview

I started coding before fully understanding the question and was not selected. The experience taught me to listen, clarify the problem, and only then solve it.

Returned

College · Teaching juniors

When my Scaler teammates left for internships, I returned to college and helped juniors and Programming Pathshala students prepare for placements.

Joined

HackerEarth · Problem Curator

With Amazon starting in July, HackerEarth became a six-month final-semester bridge into full-time engineering.

Achievements

All college achievements—earned by choosing harder challenges.

Competitive programming gave me a steady supply of difficult problems. The rankings mattered, but the lasting value was learning to stay with a challenge until I understood it.

What stayed with me

Confidence moves you forward. Humility and gratitude keep it honest.

Feedback

What mentors and peers noticed.

2 recommendations
Next

HackerEarth turned problem solving into a responsibility for fair hiring.

Designing for fairness

Jan 2020 — Jun 2020 · Bengaluru

6 months

HackerEarth
HackerEarthProblem Curator Intern

Designing hiring problems taught me to think about fairness.

A difficult question is not automatically a useful one. Clear wording, sensible limits, and complete tests shape every candidate’s chance to succeed.

MilestonesSelected outcomes
6+Company hiring programs supported
120+People at an algorithms workshop
10+ hrsWorkshop delivered over two days
Problem setting

I moved from solving problems to designing fair hiring assessments.

At HackerEarth, difficulty alone was never the goal. Every problem had to challenge candidates while remaining clear, consistent, and focused on the skill it was intended to measure.

Learnings
01

Fairness in hiring

Each problem needed clear wording, sound constraints, a correct solution, and thorough tests. The assessment had to measure skill consistently and give every candidate a fair opportunity to demonstrate what they knew.

02

Problem setting strengthened critical thinking

Creating problems pushed me to explore edge cases, design challenging variations, anticipate different approaches, and test solutions from a candidate’s perspective. Explaining these ideas further strengthened my knowledge.

Explore HackerEarth work details
Full details

Problem Curator Intern

Period and location
Jan 2020 to Jun 2020, Bengaluru
Duration
6 months
  • 01

    Curated and tested algorithmic problems for hiring and internal coding contests conducted for Google, Facebook, Nokia, PayPal, Salesforce, and Infosys, including the nationwide HackWithInfy contest with 167,000+ participants.

  • 02

    Helped increase Nokia’s post-contest Net Promoter Score from 7 to 8 by strengthening problem quality, promptly addressing participant queries and accelerated issue resolution.

  • 03

    Presented a 2-day / 10+ hour advanced data structures and algorithms workshop at MNIT Jaipur to 120+ attendees.

Technologies

C++PythonJavaJavaScriptData StructuresAlgorithms

Key skills

Problem SettingCompetitive ProgrammingTeachingPublic Speaking

What stayed with me

Expertise becomes useful when it is clear, fair, and easy to explain.

Feedback

What people at HackerEarth noticed.

2 recommendations
Next

Amazon added responsibility for software used in daily financial work.

Owning real outcomes

Jul 2020 — Aug 2021 · Hyderabad

1 year 2 months

Amazon
AmazonSoftware Development Engineer I

Amazon showed me what software means when people depend on it every day.

I worked on finance systems handling more than 100,000 invoice requests a day. The goal was simple: protect money, surface problems sooner, and keep every release dependable.

MilestonesSelected outcomes
$100M+Unusual invoices identified in one month
100K+Invoice requests handled each day
24h → liveDelay reduced to near real time
Production responsibility

Production made the customer benefit visible.

A small mistake could delay a payment or hide a costly invoice problem. Good work meant understanding the full customer need, releasing carefully, watching results, and responding quickly when something failed.

Impact highlights
01

Protect real money

I built three checks for unusual, duplicate, and potentially fraudulent invoices. They identified more than $100M in anomalies within one month.

02

Show problems while action still helps

I helped reduce the wait for anomaly reports from 24 hours to near real time, so finance teams could act sooner.

03

Make every release dependable

I added service interfaces, dashboards, alerts, automated infrastructure, and release checks so the benefit could continue safely after launch.

Explore Amazon work details
Full details

Software Development Engineer I

Period and location
Jul 2020 to Aug 2021, Hyderabad
Duration
1 year 2 months
  • 01

    Built statistical and ML-powered systems in Amazon Finance Automation to detect invoice fraud, duplicates, and anomalies at scale.

  • 02

    Built duplicate and anomaly-detection capabilities for invoice creation and update workflows handling 1 lakh+ requests every day across Amazon’s retail, non-retail, and corporate finance systems.

  • 03

    Implemented 3 anomaly-detection rules using statistical analysis and machine learning, including Isolation Forest. The rules identified over 100M USD in anomalies within the first month.

  • 04

    Reduced anomaly reporting latency from 24 hours to approximately 5 minutes through Gandalf-Veritas integration, giving finance teams near-real-time visibility into suspicious activity.

  • 05

    Built APIs and operational dashboards for monitoring and alerting, and used AWS CDK to provision infrastructure and CI/CD pipelines across multiple services.

  • 06

    Developed microservices using Java, Kotlin, TypeScript, Python, Elasticsearch, DynamoDB, Coral, Smithy, and AWS.

Technologies

JavaKotlinTypeScriptPythonAWS CDKElasticsearchDynamoDBRedshiftCoralSmithy

Key skills

Machine LearningAnomaly DetectionMicroservicesAPI DesignCI/CDMonitoring

Explore decisions and interview outcomes
Selected

Google · Second interview

A recruiter returned after my college rejection. I asked for one month, completed more than 450 practice questions, and cleared the process by listening, clarifying, and then solving.

Moved

Leaving Amazon

I was on a strong path toward SDE II, but my manager was moving and a senior engineer encouraged the wider learning opportunity. I chose Search at Google.

What stayed with me

Customer obsession: solve the real customer problem, make the benefit visible, and keep the system dependable every day.

Feedback

What people at Amazon noticed.

1 recommendation
Next

Google widened the work from finance systems to Search across languages and regions.

Understanding people at scale

Sep 2021 — Feb 2024 · Bengaluru · Hybrid

2 years 6 months

Google
GoogleSoftware Engineer II · Search India

Google taught me to connect scale with human intent.

Search work began with a simple question: what is this person trying to do? Language, local context, and careful releases all followed from that.

MilestonesSelected outcomes
1M+Estimated daily searches touched
3Indian languages launched
50+Mentoring sessions outside the role
Search at scale

The lesson from my first interview became a daily working habit.

At Google, I learned to begin by listening, clarifying the need, and understanding the people affected. That approach guided both product decisions and work across several teams.

Product learnings
01

Intent mattered more than exact words

I built education features and helped Search better understand what people wanted from exam-related questions. The change reached 0.05% of Search traffic—an estimated 1M+ daily queries.

02

Language support became shared work

I helped launch exam results in Hindi, Tamil, and Telugu, then built a common service so several features could reuse the same support.

03

Influence replaced direct control

Progress depended on clear requirements, shared priorities, patient discussion, and careful changes across several teams.

04

Mentoring kept scale personal

Across 50+ Bosscoder sessions, I supported engineers who later joined Atlassian, Adobe, Amazon, PhonePe, and Microsoft.

Explore Google work details
Full details

Software Engineer II

Period and location
Sep 2021 to Feb 2024, Bengaluru (Hybrid)
Duration
2 years 6 months
  • 01

    As part of Google’s Search India Team, built exam OSRPs and the interactive Exam Quiz, reaching 100M+ educational users on a product serving 1B+ queries per day.

  • 02

    Migrated Google Search’s serving query-understanding architecture from entity-driven to intent-driven through the diOSRP Migration, enabling better Organized Search Result Pages (OSRPs) that answered user intent directly instead of routing users to the traditional 10 blue links; impacted 0.05% of Search traffic.

  • 03

    Launched localized Exam OSRPs in Hindi, Tamil, and Telugu, adding exam dates, fees and waivers, tabbed result layouts, and the Periodic Table. Localization reached 0.03% of Search queries and 0.0019% of educational queries, improving accessibility; also designed a scalable Language API for Indic locales.

  • 04

    Contributed across Google Search’s monorepo, one of the world’s largest codebases with 2B+ lines of code and 10M+ files, using C++, Java, Python, Go, gRPC, Protocol Buffers, Knowledge Graphs, and other internal technologies.

  • 05

    Collaborated across Search teams to resolve production issues, align requirements, and prioritize high-impact improvements.

  • 06

    Contributed to Google Bard’s early model-improvement journey by training and evaluating LLM responses, helping improve output quality as the technology evolved from unreliable early behavior toward practical usefulness.

Technologies

C++JavaPythonGogRPCProtocol BuffersGraph Database (Knowledge Graph)Monorepo

Key skills

Product EngineeringAPI DesignLocalizationCross-functional CollaborationSystem Design

Explore decisions and interview outcomes
Life

Marriage

I married on 28 November 2023. Career, location, and daily life now belonged in one decision.

Ambition

Wider hands-on ownership

After two and a half years, I wanted responsibility from design and implementation through delivery and team growth.

Rejected

Amazon · SDE II

I gave the interview without any preparation and was not selected.

Selected

Oracle Health

Healthcare impact, remote work, and broader ownership led me to join in April 2024.

What stayed with me

Preparation starts as an unclear problem. Begin with the time available, decide what matters most, and follow a clear path—just as with any engineering problem.

Next

Oracle Health added wider ownership across product delivery and team growth.

Growing systems and people

Apr 2024 — Present · Oracle Health · Remote

Tech Lead since Oct 2025

Oracle
OracleSenior Software Engineer · Tech Lead

Oracle brought product delivery and team growth into one role.

I now lead work across design, delivery, hiring, and team growth for a healthcare reporting product where reliability directly affects people’s work.

MilestonesSelected outcomes
10+Engineers across three products
7+Services designed and built
100+Technical interviews conducted
Leadership shift

I chose broader ownership and work that fit the life we were building.

Oracle Health offered meaningful problems, room to lead, and a flexible working model. The role grew from building software to helping a team deliver it well.

Leadership highlights
01

Make clinical reports dependable

I guide the reporting product across requests, templates, data, final documents, security, history, and recovery when work fails.

02

Fix the limit behind the failure

For very large reports, the team cut memory use by 40%, improved speed by more than 25%, and reduced repeated data work by 60%.

03

Grow people, not dependency

I built the India team, led four quarterly releases, partnered with 25+ teams, and helped make 25+ hires while keeping the product from depending on one person.

Explore Oracle work details
Full details

Senior Member of Technical Staff (SDE III), Tech Lead

Period and location
Apr 2024 to Present, Bengaluru (Remote)
Duration
Promoted to Tech Lead in Oct 2025
  • 01

    Technical Lead for Oracle Health’s Clinical Reporting / Output Management team, owning the architecture, technical direction, delivery, and production readiness across 3 products: MRO Gen2, XR Gen1, and Dex Gen1.

  • 02

    Built 7+ Java Micronaut microservices from 0 to 1, transforming a medical-record output platform into a modular Gen2 reporting system spanning APIs, template management, data extraction, report generation, and the mro-ai-intelligence-service. Proposed modern AI capabilities, including an advanced AI-powered template builder, shaped the product roadmap, and aligned execution with product managers and partner teams.

  • 03

    Scaled the platform to handle 1M+ requests per day while keeping p90 latency under 30 seconds. For the roughly 1% of complex reports that take more than 10 minutes, and sometimes hours, introduced asynchronous fast and slow lanes with bulk generation, retries, DLQs, idempotency, horizontal scaling, security, dashboards, and alerting.

  • 04

    Led product delivery across 4 quarterly releases, from roadmap planning and architecture through production launch. Broke down work for 5 to 8 engineers, removed blockers, and drove architecture decisions, defect resolution, compliance, monitoring, and operational readiness.

  • 05

    Served as a primary escalation owner for critical production issues, leading high-severity investigations through diagnosis, remediation, and prevention. In one example, brought approximately 1,000-page reports back within the 30-second SLA, reducing heap usage by 40%, improving performance by over 25%, and driving request-level caching that cut data-extraction time by 60%.

  • 06

    Partnered with 25+ cross-functional teams to turn custom clinical-reporting needs into reusable platform capabilities. Cut Gen2 onboarding from 2 weeks to under 1 week, a reduction of more than 50%.

  • 07

    Led a project involving 3 engineers to deliver Gen1 Report Ingress and 6 APIs in 1 quarter + 1 month versus 2 quarters planned; enabled WebSphere-to-EJS migration with better performance, stability, scalability, and lower third-party cost.

  • 08

    Took ownership of highly ambiguous technical and product scopes, turning open questions into clear architecture and execution plans across multiple initiatives, including early-stage planning for the SI DRZ region build and EHRC modernization program.

  • 09

    Built and led the India Clinical Reporting team from scratch, growing the leadership scope to 10+ engineers across 3+ products. Strengthened organization-wide hiring as a Senior Key Interviewer (Bar Raiser), conducting 100+ interviews and contributing to 25+ hires; also mentored and judged 12 teams developing AI projects at Oracle-wide OraHacks hackathons.

Technologies

JavaMicronautReactKafkaRedisSQLElasticsearchLarge Semantic Object StorageOracle Cloud InfrastructureMicroservices

Key skills

Technical LeadershipSystem ArchitecturePerformance TuningHiringMentoringCross-team Delivery

Explore decisions and interview outcomes
Rejected

Meta

I interviewed after joining Oracle and was not selected.

Two offers

Amazon · SDE II

Two separate offline hiring drives produced two offers.

Offer

Uber · SDE II

I was selected through an offline hiring drive.

Offer

Microsoft · L62

I stayed at Oracle because ownership, growth, and impact mattered more than changing logos.

Rejected

xFlow

I interviewed with the startup but was not selected because the cofounders and I had differing opinions.

Rejected · First round

Uber · Senior Software Engineer (L5A)

Hard luck · bad day.

What stayed with me

Leadership creates clarity, grows judgement, and leaves the team stronger than one individual.

Feedback

What people at Oracle noticed.

4 recommendations
Looking ahead

The next role should deepen product ownership, people leadership, and long-term impact together.

What the journey adds up to

The next step is about responsibility, not a logo.

College built confidence. HackerEarth added fairness. Amazon added responsibility. Google added empathy at scale. Oracle brought product delivery and team growth together.

Explore the portfolio