Entering the orbit of

Builder · Competitor · Problem solver

Aligning the starsLoading achievementsReady

Senior AI Engineer · Citi

Vaibhav
Patel

Always a builder

I build production AI and backend systems at Citi, and I win the rooms that are hardest to win.

Citi PIR 2026of 100+ projects
#1
IDFC FAME 2.02,000+ competitors
₹5L
In two yearsthe usual path: five
C9→C11
Codeforces Expertpeak rating
1627
Scroll to begin the journey
  • Citi PIR 2026 · #1 of 100+
  • IDFC FAME 2.0 · ₹5,00,000
  • Citi GenAI Prompt Engineering · #1
  • C9 → C11 in two years
  • Codeforces Round 925 · AIR 19
  • Codeforces Expert · 1627
  • CodeChef 5★ · 2007
  • Meta Hacker Cup 2024 · Round 2
  • ICPC Asia West Regionalist
  • JEE Advanced · AIR 6913
  • MHT CET · AMR 50
  • Selected for Google Summer of Code 2024

Where I've built

  • NowCitiSenior AI Engineer2024 – now
  • Goldman SachsSummer Analyst2023
  • Google Summer of CodeSelected contributor2024
  • Internet ArchiveOpenLibrary2024
  • VJTI MumbaiB.Tech, Computer Engg.2021 – 24

Hall of achievements

Rank is a habit.

From competition floors to production systems: the results that define the climb.

Flagship · 2026

#1

of 100+ projects

Citi PIR 2026 · Winner

Ranked first among every Citi 2024 graduate with RUNNEX, an API workflow and load-testing platform now used by 200+ people.

APIs
227
users
200+
runs / month
2,000+
Read the RUNNEX story
International · ML · Nov 2023

₹5,00,000

2,000+ participants · one winning team

IDFC FIRST Bank FAME 2.0 · Winner

An international machine-learning hackathon, won on 3 November 2023 while I was still at VJTI. The prize came on a cheque as big as a banner.

  • #1Citi PIR 2026 · Winner
  • ₹5,00,000IDFC FIRST Bank FAME 2.0 · Winner
  • #1Citi GenAI Prompt Engineering · Winner
  • C9 → C11Senior in two years
  • AIR 19Codeforces Round 925 (Div. 3)
  • 1627Codeforces Expert
  • 2007CodeChef 5★
  • AIR 545Meta Hacker Cup 2024 · Round 2
  • ICPCICPC Asia West Regionalist
  • AIR 6913IIT JEE Advanced
  • AMR 50MHT CET
  • TopTop inner-sourced project at Citi

Every star is a result. Together they tell one story.

Every achievement, with proof

The journey

Two years. Three levels.

C9 to Senior AI Engineer in about two years. The usual path takes five.

  1. 202001 / 14 · Foundations

    AMR 50

    MHT CET

    99.97 percentile in Maharashtra's engineering entrance exam.

  2. 202102 / 14 · Foundations

    AIR 6913

    IIT JEE Advanced

    Cleared India's toughest engineering entrance exam.

  3. Jan 202103 / 14 · VJTI · B.Tech

    VJTI Mumbai

    B.Tech in Computer Engineering. The batch started in January because of COVID-19.

  4. Jun 202304 / 14 · VJTI · B.Tech

    Goldman Sachs · Summer Analyst

    Shipped a full-stack fee-schedule app in Asset & Wealth Management, with 96% backend test coverage.

  5. Nov 202305 / 14 · VJTI · B.Tech

    ₹5,00,000

    IDFC FIRST Bank FAME 2.0 · Winner

    An international ML hackathon with 2,000+ participants. Won it while still in college.

  6. Jan 202406 / 14 · VJTI · B.Tech

    ICPC Asia West Regionals

    Qualified for Amritapuri, a month after becoming a Codeforces Specialist.

  7. Feb 202407 / 14 · VJTI · B.Tech

    AIR 19

    Codeforces Round 925

    Global 131 of 42,000+. Expert rating followed.

  8. Jun 202408 / 14 · VJTI · B.Tech

    Graduated · GSoC

    B.Tech with a CGPA of 8.25. Selected for Google Summer of Code with Internet Archive.

  9. Jul 202409 / 14 · Citi · C9

    Joined Citi · C9

    Software Engineer in Pune. Built PII detection, bulk onboarding (−70% time) and a top inner-sourced data platform.

  10. Oct 202410 / 14 · Citi · C9

    AIR 545

    Meta Hacker Cup · Round 2

    Global rank 2254 in Meta's worldwide contest.

  11. Feb 202511 / 14 · Citi · C9

    #1

    Citi GenAI Prompt Engineering

    First of 100+ participants.

  12. Jul 202512 / 14 · C10

    Promoted · C10

    GenAI document platforms, an agentic compliance scanner and a RAG onboarding assistant.

  13. Sep 202613 / 14 · C11 · Senior

    Promoted · C11

    Senior AI Engineer. Client Insights AI agent and a GenAI CRM summariser.

  14. 202614 / 14 · C11 · Senior

    #1

    Citi PIR 2026 · Winner

    RUNNEX ranked first of 100+ projects across every 2024 graduate.

2 yrsC9 → C11, Vaibhav
~5 yrsC9 → C11, typical

Systems built

Built to be used. Built to last.

Production platforms, GenAI pipelines and agentic systems, each one shipped to real users with real stakes.

All systems

Citi · 2026

PIR 2026 · #1

RUNNEX

API workflow & load-testing platform

Problem
Testing real API journeys meant hand-written scripts scattered across teams.
Built
227 Java (Spring Boot) APIs over 50+ data models with 500+ fields · A React interface assembled from 90+ reusable UI components
Impact
#1 Citi PIR 2026, of 100+ projects200+ users2,000+ executions every month
  • Java
  • Spring Boot
  • React
  • REST
Case study

Citi · 2025–26

Data Controls Scanner

Agentic AI for compliance discovery

Problem
Compliance needed a map of the data elements and controls that live across many repositories.
Built
An agentic AI pipeline that reads codebases and identifies data elements and controls · A metadata scanner that filters and batches files so large repositories stay tractable
Impact
90+ FastAPI servicesMulti-role RBAC review workflowAutomated replacing manual audits
  • Python
  • FastAPI
  • Agentic AI
  • LLMs
Case study

Citi · 2025

GenAI PDF Platform

Automated PDF tagging & PDF Analyzer

Problem
Preparing PDF forms (Acroforms) for prefilling meant tagging every field by hand.
Built
A GenAI system that tags PDF Acroform fields automatically · A preview UI to validate every suggested tag against the document
Impact
Eliminated manual tagging workflows40–45% less manual effort in tag editing
  • Python
  • GenAI
  • Prompt engineering
  • React
Case study

Citi · 2026

Client Insights AI Agent

Conversational AI with entitlements, plus a CRM summariser

Problem
Sales and admin teams needed fast answers about their clients, and their meeting notes were unstructured text nobody could analyse at scale.
Built
Chat with session history for GenAI conversations on production data · Entitlement-aware results scoped to each user's managed segments
Impact
20 outcome classes for call notesRole-aware answers by managed segment
  • Python
  • GenAI
  • LLMs
  • RBAC
Case study

Citi · 2024–25

PII Detection in Logs

NLP + GenAI compliance monitoring

Problem
Spotting personal data in application logs by hand was unreliable and couldn't scale with the volume of logs being produced.
Built
NLP models (spaCy) for tokenisation, stemming and entity recognition · GenAI-assisted detection to catch what rules and models miss
Impact
Automated compliance checks on logsNLP + GenAI layered detection
  • Python
  • spaCy
  • GenAI
  • Next.js
Case study

Goldman Sachs · 2023

Fee Schedule Capture

Full-stack app at Goldman Sachs

Problem
Negotiated investment-management fees for special investors needed a reliable, auditable way to be captured.
Built
Spring Boot REST APIs for CRUD over a NoSQL store · A responsive React UI, modularised to cut compilation time by 30%
Impact
96% backend test coverage82% frontend test coverage−30% compile time
  • Java
  • Spring Boot
  • React
  • TypeScript
Case study

The arena

Where the thinking gets sharpened.

Contests are the gym: timed, ranked and unforgiving. They're the reason production problems feel solvable.

Codeforces Expert · peak 1627
  • 1627

    Codeforces Expert

    peak rating

  • 2007

    CodeChef 5★

    peak rating

  • AIR 19

    Codeforces Round 925

    Global 131 of 42,000+

  • AIR 545

    Meta Hacker Cup 2024

    Round 2 · Global 2254

  • ICPC Asia West Amritapuri Regionalist · 2023–24
  • Codeforces Round 909 · Rank 522 of 10,000+ → Specialist
  • CodeChef Starters 113 · Rank 128 (Div 2)

The arsenal

Everything orbits problem solving.

Languages and frameworks change. The core doesn't.

At the core: Data structures & algorithms · System design · Competitive programming

Agentic pipelines, RAG and GenAI that run in production.

  • Agentic AI · LangGraphAgentic compliance pipelines
  • LangChainGenAI platforms
  • RAGREST Onboarding Assistant
  • Prompt engineering#1 at Citi GenAI competition
  • LLMs · GPT-4PII detection, PDF tagging
  • spaCy · NERPII detection
  • scikit-learnML hackathons
  • Pandas · NumPyData & ML work

Beyond the code

Always a warrior.

A perfectionist by habit. I solve problems by understanding the people inside them, and I read history to understand why the world works the way it does. I'm always after the truth.

  • Own it end to end.
  • Measure, don't guess.
  • Never settle for almost.
  • Chess

    Patience first, then precision. Played since school.

  • Table tennis

    Reflexes and reading spin. A junior-college favourite.

  • Cricket

    A team sport and a long game.

  • Badminton

    Speed, footwork and no wasted movement.

  • Mathematics

    My first love, and the reason competitive programming clicked.

  • World history

    How empires rise and fall teaches more about systems than most textbooks.

Drag to turn · click to move

Proof

It happened. Here's the evidence.

Credentials, certificates and the moments behind the numbers.

What's next

Let's build something that wins.

I'm always up for a conversation about hard problems, AI systems at scale and teams that play to win.