Projects

Real systems.
Real data. Real impact.

Production-style work across analytics engineering, AI automation, data platforms, quantitative systems, and the Unwind AI product ecosystem — from ingestion to insight delivery.

Unwind AI Ecosystem

Flagship AI venture
and its products.

Four interconnected products built on a common AI automation architecture. Open-weight model infrastructure, retrieval workflows, domain orchestration, and product-embedded intelligence.

Venture
AI Venture · Open-weight Automation Layer

Unwind AI

A custom AI automation venture building production-grade intelligence systems using open-weight models, retrieval architectures, prompt layers, guardrails, APIs, and workflow integrations. The umbrella platform powering KYC Agri, Happy Agri, and Unwind Dreams.

Intelligence architecture

Business Data → Retrieval (RAG / Vector Search) → Prompt Layer → Model Layer (Llama / OSS) → Guardrails → API Actions → Product Output

Core capabilities

Open-weight model workflows · Llama-powered local AI · RAG and vector search · Workflow automation · API-connected agents · Private / local deployment · Product-embedded AI · Decision-support systems

Open-weight LLMs RAG Vector Search Workflow Automation Local Inference Python API Integration
Visit Unwind AI ↗ Discuss a Build
Live Product
Agricultural Intelligence · Commodity AI

KYC Agri

An agricultural and commodity intelligence platform where Unwind AI acts as the intelligence layer. Supports article summarization, market signal extraction, commodity research workflows, and AI-assisted price prediction. Built for agri traders, commodity professionals, and market researchers.

Key capabilities

Commodity intelligence · Article summarization · Market signal extraction · Predictor support · Premium AI features · Research acceleration · Local / private AI backend possibilities

Commodity AI Article Summarization Market Signals RAG Price Prediction
View KYC Agri ↗
🌾
In Development
AgriTech Decision Support

Happy Agri

An AI-powered agri decision-support platform converting farm inputs, crop history, weather context, and market data into practical recommendations for farmers, traders, and agri businesses.

Decision engine layers

Farm Inputs → Retrieval Layer → Weather + Market Context → AI Advisory → Recommendations → Action Plans

Decision Engine Crop Intelligence Market Data Advisory AI Knowledge Retrieval
View Concept
🌱
Coming Soon
iOS App · AI Goal Coach

Unwind Dreams

A premium AI-powered goal achievement app. Helps users talk through their ambitions, convert them into realistic plans, track daily execution, adapt routines, and stay accountable through a social layer. iOS-first. Premium personal operating system feel.

Core experience

Conversational goal-setting · Adaptive daily plans · Progress dashboards · Streak tracking · Friends accountability · AI plan adaptation · iOS-first · Premium personal OS feel

iOS App Conversational AI Goal Coaching Adaptive Planning Social Layer
Preview Concept Join Waitlist

Engineering Projects

From raw input to
decision-ready systems.

Across the strongest projects, the core pattern stays consistent: source data is collected, validated, transformed, modeled, and exposed through dashboards, analytics applications, or domain outputs.

Flagship Project

Formula One
Analytics Platform

History Covered
75 Years of F1 Data
Architecture
5-Layer Pipeline
Core Stack
Python · Spark · Snowflake

A full analytics platform designed around the complete data lifecycle: raw ingestion, distributed transformation, warehouse-ready modeling, and insight delivery through dashboards and a live Streamlit application. Platform thinking — not just analysis.

End-to-end architecture

Source datasets → Python ingestion → Spark / Databricks transformation → cleaned analytical tables → warehouse modeling → dashboards + Streamlit application

How it works

Sources
Race & Historical Data

Race results, driver records, constructors, lap data, qualifying, season history.

Ingestion
Python ETL

Raw inputs collected, validated, typed, and standardized before downstream processing.

Processing
Spark / Databricks

Distributed transforms clean, normalize, enrich historical records into reliable layers.

Warehouse
Structured Modeling

Analytics-ready tables and marts support comparisons, standings, trend analysis.

Consumption
BI + Live Demo

Dashboards and Streamlit expose the platform to analysts and stakeholders.

Python Apache Spark Databricks Snowflake R Looker Studio ETL Data Modeling Streamlit

Live Interactive Demo

Interactive F1 Analytics Platform

Explore end to end — ingestion, Spark transforms, warehouse model, BI reporting, driver analytics, lap-time performance.

🏎
Analytics Engineering

GA4 Analytics
Dashboard Pipeline

An end-to-end analytics pipeline transforming raw GA4 event data into structured, decision-ready dashboards. Event collection, metric logic, transformation layers, and reporting outputs.

End-to-end architecture

GA4 event source → Python extraction → cleaning + metric transformation → analytics-ready tables → Looker Studio dashboard

GA4 Python Looker Studio SQL Analytics Engineering
View on GitHub
📈
Applied AI

AI Studio —
RL Environment

A reinforcement learning experimentation environment for training agents under custom reward structures, environment rules, and iterative training workflows.

End-to-end architecture

Environment design → state representation → reward logic → agent training loop → evaluation runs → performance iteration

PyTorch Reinforcement Learning Python TensorFlow Experimentation
View on GitHub
🧠
Quantitative Systems

High-Frequency
Trading Simulation

A Python-based simulation system for testing trading behavior in synthetic market conditions — rule-based decisions, timing, and performance analysis.

End-to-end architecture

Synthetic market generator → pricing / signal logic → trade execution simulation → latency-aware evaluation → strategy performance analysis

Python Quantitative Modeling Algorithm Design NumPy Simulation
View on GitHub
📉
Big Data

Movie Data Platform
on Databricks

Distributed data processing built on Databricks and Apache Spark — ingestion, transformation, cleaning, and large-scale analysis of movie datasets.

End-to-end architecture

Raw movie datasets → Databricks ingestion → PySpark transformation → cleaned distributed layers → SQL analysis / reporting

Databricks Apache Spark PySpark SQL Big Data
View on GitHub
🎬
Data Ingestion

Automated
Data Mining System

A structured scraping and ingestion pipeline for collecting, validating, and preparing data from external sources for downstream analytics.

End-to-end architecture

External web sources → scraping layer → parsing + validation → cleaned structured records → analytics-ready datasets

Python Web Scraping Data Pipelines Automation Data Validation
View on GitHub
🕸