Abhijeet KaduSDE & PM
Mumbai, IndiaPortfolio 2026
000
Loading
Abhijeet Kadu
Open to Work
Mumbai, IN2026SDE + PM

Abhijeet Kadu
builds things.

SDE · Product Manager · ML Engineer.
Full-stack from Jupyter notebooks to production deployments.
Mumbai-based, globally useful.

“Stop guessing what good engineers look like. Here’s one.”

01

The Generalist Trap

Recruiters see "SDE & Product Manager" and assume you’re a diluted version of both. You’re actually just rare.

02

The Silent Rejection

A portfolio without live projects is a resume in disguise. Every repo here ships.

03

The Experience Gap

50K resumes scored. 150K ML records processed. Top 0.1% nationally. The gap is already closed.

FastAPINext.jsAWSGCPXGBoostPostgreSQLTypeScriptDockerRAGVertex AIFAISSSocket.ioLangChainOAuth2CI/CDPythonFastAPINext.jsAWSGCPXGBoostPostgreSQLTypeScriptDockerRAGVertex AIFAISSSocket.ioLangChainOAuth2CI/CDPython
Building systems that scaleOpen to 2026 rolesMumbai · IndiaTop 0.1% · Amazon ML50K+ resumes processed99.9% cloud uptimeSDE & Product ManagerAvailable June 2026Building systems that scaleOpen to 2026 rolesMumbai · IndiaTop 0.1% · Amazon ML50K+ resumes processed99.9% cloud uptimeSDE & Product ManagerAvailable June 2026

01 / Building

Building
systems
that ship.
CurrentlyEcovis RKCA — SDE & PM
Graduating2026 · Computer Engineering
CertifiedDP-600 · Microsoft Fabric
SelectedAmazon ML Summer School
LocationMumbai, IN · Open to remote
SeekingSDE / ML / Full-stack roles
Available June 2026

I’ve been writing code that runs in production since my second year — a room-finder with live users, a resume engine scoring 50,000+ CVs, an OCR finance system processing receipts in real time.

Right now I’m at Ecovis RKCA: migrating legacy infra to AWS, building a RAG assistant on GCP Vertex AI, and acting as PM on a client-facing product — all at once.

I graduate in 2026. Looking for a backend, ML, or full-stack role where I can own something end-to-end from day one. I don’t wait to be assigned work — I find what’s broken and fix it.

Amazon ML School · 2024

Top 0.0%

of 100,000+ applicants nationally — selected for Amazon’s ML Summer School program

Status

Available

June 2026 · Full-time roles · Open to remote & relocation

Cloud Uptime

0%

post-migration at Ecovis RKCA (AWS)

Resumes Scored

0K+

via SmartResume ATS engine

Reporting Cut

0%

via Next.js internal tooling at Ecovis

Shipped

0+

products end-to-end in production

Tech Stack Snapshot

Next.jsFastAPIAWSGCPPythonTypeScriptPostgreSQLDockerRAGXGBoostPyTorchMS Fabric

02 / Where

Where
I've
shipped.
Current

Software Development Engineer & Product Manager

Ecovis RKCA

2026 – PresentFull-Time
  • Migrated legacy infrastructure to AWS & Azure — improved uptime to 99.9%
  • Led product roadmap end-to-end: requirements, sprint planning, and delivery across cross-functional teams
  • Built internal Next.js + TypeScript tooling that reduced manual reporting time by 60%
  • Designed CI/CD pipelines that cut release cycles from two weeks to two days
  • Defined KPIs with stakeholders and drove data-informed product decisions
Microsoft DP-600

Microsoft Fabric Data Engineering Intern

Microsoft

2026Internship
  • Built Bronze → Silver → Gold ETL medallion pipelines on Microsoft Fabric
  • Created Power BI semantic models and executive dashboards
  • Earned DP-600 Fabric Analytics Engineer certification
top 0.1%

ML Competition Finalist

Amazon ML Summer School

2025Achievement
  • Top 0.1% nationally among 100,000+ applicants
  • Ensemble of XGBoost + LightGBM + CatBoost with SMOTE oversampling
  • Selected for Amazon ML Summer School program

04 / Selected

Selected
projects
7 projects
01

Personal Project · 2025

featured

SmartResume — Resume Analysis Engine

ATS scoring engine that processes and scores 50,000+ resumes.

FastAPI backend with XGBoost classifier. Feature extraction from structure, keyword density, and section completeness. Secure OAuth2 + JWT auth.

FastAPIXGBoostPythonOAuth2JWTPostgreSQL

Resumes

Production

02

Personal Project · 2024

Room & Food Finder

Live hyperlocal platform — students use it daily to find housing and food near campus.

Multi-role system (User/Provider/Admin) with real-time chat via Socket.io. Map-centric geo-search, Dockerized services, automated CI/CD.

Node.jsSocket.ioNext.jsMongoDBDockerMaps APIRBAC

Active users

Real-time

03

Ecovis RKCA · GCP · 2026

RAG Knowledge Assistant

Internal knowledge retrieval — ask a question, get cited answers in seconds.

Doc ingestion → chunking → FAISS vector store → LLM answer generation. Deployed on GCP Vertex AI with streaming responses via SSE.

RAGVertex AIFAISSLangChainGCPPythonLLMs

Deployed

Streaming

04

Amazon ML Summer School · 2025

top 0.1%

Amazon ML — Product Price Prediction

End-to-end regression on 150,000+ records. Top 0.1% nationally.

EDA → feature engineering → XGBoost → hyperparameter optimization. Ensemble approach with bias-variance analysis.

XGBoostPythonFeature Eng.PandasScikit-learnEDA

Records

National

05

Personal Project · 2025

Marketing Analytics Dashboard

Power BI dashboard for CMOs — 5,000+ campaign records → actionable ROI.

Star schema data model with advanced DAX. ROAS, CPA, MoM growth. Interactive drill-through across Facebook, Google Ads, LinkedIn.

Power BIDAXData ModelingStar SchemaROASCPA

Campaigns

Ad channels

06

Personal Project · 2025

Medical Triage RL Environment

RL environment simulating emergency room triage decisions under constraint.

Custom OpenAI Gym-compatible environment modeling patient arrival, severity scoring, and resource allocation. Agents learn via Q-learning and PPO.

Reinforcement LearningPythonOpenAI GymPPOQ-LearningHealthcare AI

Policy

Compatible

07

Personal Project · 2025

SalesVision — Sales Analytics

Full-stack sales intelligence with live dashboards and forecasting.

Real-time sales metrics, trend forecasting, and team performance tracking. Built for sales managers to cut through noise and act on signal.

JavaScriptAnalyticsDashboardCharts.jsREST APIForecasting

Dashboard

Updates

03 / Skills

Skills
&
expertise

Cloud & Infra

AWS (EC2 · S3 · IAM · CloudWatch)GCP · Vertex AIDP-600 CertifiedMicrosoft FabricDockerCI/CD · GitHub ActionsAzure

Backend & Systems

Node.jsFastAPIREST APIsSocket.ioServer-Sent EventsOAuth2 · JWTRBACPostgreSQLMongoDB

ML & Data

RAG · LLMsXGBoostFAISS Vector StoreLightGBMCatBoostPandas · NumPyScikit-learnPower BI · DAX

Frontend & Product

Next.jsReactTailwind CSSGSAPTypeScriptSprint PlanningKPI DefinitionRoadmapping

Certifications & Recognition

DP-600

Fabric Analytics Engineer

Microsoft · 2024

Top 0.1%

Amazon ML Summer School

Amazon · 2024

Patent 458179-001

Smart Inventory System

Govt. of India · 2024

Interactive

Which kind of engineer
are you?

Pick 3 technologies you work with (or want to). I’ll tell you what kind of engineer you’d be on my team.

0 / 3 selected

06 / Certifications

Certifications
& credentials
2 certificates

Amazon · 2025

ML Summer School

top pick
Supervised LearningXGBoostFeature EngineeringModel Evaluation
▼ details

Microsoft · 2026

Fabric Analytics Engineer DP-600

top pick
MS FabricSparkLakehousePower BIETL
▼ details

06 / Open

Open
Source
contributions
4 merged PRs

Real-world contributions to codebases used by millions — TensorFlow, MLflow, and OpenLightingProject. Every PR was reviewed, approved, and merged by core maintainers.

merged
Fix typo in XLA constants

Corrected a typo in XLA kernel constant naming — merged into main.

Merged
Update assert_no_garbage_created docstring

Improved docstring clarity for the garbage-collection test helper.

Merged
mlflow
mlflow/mlflow

1 merged PR

merged
Fix typo in PromoteModelButton

Fixed a UI label typo in the Model Registry promote button.

Merged
Improve LabeledInput component

Enhanced the LabeledInput UI component with better accessibility.

Merged

07 / Contact

Don’t beshy about it.Let’s talk.

Open to opportunities

Mumbai, IN  ·  Available 2026

© 2026 Abhijeet Kadu