Available · Product Analytics / Applied ML · Remote

Raj Jaiswal

I improve retention, forecasting, and product decisions using ML + analytics.

Built real-world analytics systems with measurable impact — retention, forecasting, and growth.

End-to-end ML systems — from raw data to dashboards to decisions. Focused on outcomes that drive action.

See Case Studies GitHub
Churn AUC
0.84
↑ from 0.71 · Telecom 50K users
Forecast MAPE
8.3%
↓ from 14.1% · Retail sales
Conversion Lift
+18%
↑ A/B test · Checkout flow
Systems Built
4
End-to-end · Production-style
Real Work

Experience

Delivered analytics and ML systems in real contexts — freelance, open source, and production-style builds.

📈
Freelance Analytics
Client Projects · 2024–Present
  • Delivered analytics workflows and dashboards
  • Automated reporting (significant time reduction)
  • Built decision-support systems
🗄️
Data Systems
Portfolio Projects · 2024
  • Built ELT pipelines and KPI dashboards
  • Designed metrics for business tracking
  • Implemented churn detection logic
🌐
Open Source
KolibriOS · Rust · 2024
  • Contributing to OS kernel (KolibriOS, Rust)
  • Working with a global async team
  • Experience in production-level systems
Approach

How I work

ML is only useful if it leads to decisions.
I build end-to-end systems that can be measured — not one-off models in a notebook.

⚡

ML is only useful if it changes a decision

If the model doesn't lead to an action, it's a report — not a system

🔧

Systems > one-off analysis

Reproducible pipelines beat ad-hoc notebooks every time

🎯

Ship simple, measure impact

A simple model in production beats a complex one in a notebook

Case Studies

Systems I built.
Decisions they enabled.

Problem → system → metric → business decision.

Churn prediction ML dashboard showing risk scores and feature importance

Churn System for Retention Ops

Identified high-risk users early — enabling targeted retention instead of blanket campaigns.
AUC improved 0.71 → 0.84 on 50K telecom users

XGBoost SMOTE Python Feature Eng. Looker Studio
View Case Study → See full pipeline + decisions
Sales forecasting dashboard with time-series predictions and MAPE metrics

Sales Forecasting for Inventory

Reduced forecasting error to enable better inventory and resource planning decisions.
MAPE improved 14.1% → 8.3% using ARIMA + Prophet

ARIMA Prophet Python Time-Series SQL
View Case Study → See full pipeline + decisions
Synlitics restaurant analytics SaaS dashboard with KPI metrics and trends

Restaurant Analytics SaaS Pipeline

Enabled teams to track key metrics and make faster, data-driven operational decisions.
Automated ELT pipeline surfacing revenue trends & margins daily

Supabase PostgreSQL n8n Power BI ETL
View Case Study → See full pipeline + decisions
A/B testing framework dashboard showing experiment results and statistical significance

A/B Testing Framework for Checkout

Built a reusable experiment engine to validate product changes with statistical confidence.
Confirmed +18% conversion lift — a real product decision

A/B Testing Python Hypothesis Testing SQL
View Case Study → See full pipeline + decisions
Live Systems

Live Analytics Systems

Interactive dashboards built with real data — click to explore each system live in Looker Studio.

Marketing Analytics Dashboard

Marketing Performance Dashboard

Tracks impressions, clicks, conversions, and ROAS across campaigns. Identifies high-performing channels and optimizes spend.

View Live Dashboard →
Finance Dashboard

Finance & Revenue Dashboard

Revenue, costs, and profit trends over time. Highlights performance changes, category breakdowns, and geographic distribution.

View Live Dashboard →
Capabilities

What I can do for your team

Grouped by how I use them. Strong focus on learning fast and applying concepts to real datasets.

🎯 Decision Systems

  • A/B Testing — experiment design & analysis
  • SQL — cohort, funnel & retention analysis
  • Metric Design — defining KPIs that drive action
  • Statistical Inference — hypothesis testing
  • Cohort & Funnel Analysis

🧠 ML Systems

  • XGBoost — classification & ranking
  • ARIMA / Prophet — time-series forecasting
  • Feature Engineering — domain-specific signals
  • SMOTE — handling imbalanced datasets
  • Model Evaluation — AUC, precision/recall, MAPE

⚙️ Data Pipelines

  • Python — modeling & automation
  • ETL / ELT — automated data pipelines
  • PostgreSQL / Supabase — data storage
  • REST APIs — data ingestion & webhooks
  • n8n — workflow automation

📊 Dashboards & Reporting

  • Power BI
  • Looker Studio
  • Matplotlib / Seaborn
  • Automated Reporting
  • Stakeholder Communication
Trust

What Clients Say

Feedback from real projects — freelance analytics, data consulting, and collaborative work.

"

Excellent technical expertise and strong problem-solving skills. His ability to turn complex data into clear, actionable insights was impressive. He consistently delivers high-quality work on time.

Suraj K. Senior Consultant Data Analytics & Business Operations (Kivi Labs)
✔ Verified Client
"

Professional, responsive, and delivered high-quality work on time. Communication was clear and attention to detail was excellent.

Shubham G. OECD Data Analysis Project Freelancer
✔ Verified Client
"

Raj is professional, communicates clearly, and understands requirements well. Delivered quality work on time.

Asif Ali K. Q4 Sales Analysis Freelancer
✔ Verified Client
"

Reliable, communicates well, and completes tasks efficiently. I'd be happy to work with Raj again.

Client Review Project Engagement Upwork
✔ Verified Client
Let's Talk

Available for Product Analytics / Applied ML opportunity
Remote · 2026

Systems-oriented and ready to contribute from day one.

Get in Touch GitHub LinkedIn