Prathamesh Ranjane

PUNE, INDIA · AVAILABLE FOR CONTRACT & FULL-TIME

I build software that ships.

Three products built end-to-end and solo — one live in production. I use AI-assisted development to go from problem to deployed system in weeks, not quarters.

See the live product → Email me about a project
LIVE SAAS IN PRODUCTION3 PRODUCTS BUILT SOLODJANGO · REACT · SUPABASE · N8NIST — OVERLAPS US & EU

Work

● Live
grepMo dashboard showing report generation interface
grepMo detailed view of enrichment workflow
grepMo PDF export and formatting options
grepMo report customization and template management
Problem
Repetitive, manual report production — professionals spend hours reformatting raw scanner output into client-ready documents, with no standard tooling.
Built
A platform that ingests raw scanner output, parses and structures it, runs it through an editable enrichment step, and outputs a formatted professional PDF. Multi-tenant, subscription billing, zero-storage browser mode. Deployed on Railway behind Cloudflare.
Decision
Built organic discovery through AI-assisted SEO/AEO content instead of paid acquisition — the target audience searches for the problem, not the product.
Stack
Django 5Django REST FrameworkReact 18TypeScriptSupabase PostgreSQLWeasyPrintRazorpayRailwayCloudflare

StackRev

◐ Built, paused
StackRev dependency graph and architecture analysis view
StackRev impact analysis and component relationships
StackRev codebase scanning and analysis output
Problem
AI coding assistants lose the thread on unfamiliar or large codebases — token limits mean incomplete context, which produces suggestions that break dependent components.
Built
A tool that scans a codebase's structural architecture without reading business logic, producing dependency maps, impact analysis, and a compact architecture summary that can be fed to an AI assistant instead of raw files. The scanner parses Python, TypeScript/JavaScript, and C# via Tree-sitter.
Decision
Paused after identifying a distribution mismatch: the right form factor is a native IDE integration, not a standalone tool. Strategic call, not a technical failure.
Stack
Node.jsJavaScript (ES Modules)Tree-sitter (WASM)web-tree-sitterNext.js 16React 19TypeScript 5Tailwind CSS 4ESLint 9

DoxMint

○ Foundation, superseded
Problem
Repetitive manual document creation across professional workflows — inspections, audits, client onboarding — with every document produced from scratch.
Built
A questionnaire-driven document engine: conditional logic, modular template assembly, automated PDF and DOCX output, multi-tenant architecture.
Decision
Recognised the scope was too broad to solve well for any single customer. Paused it and used its architecture as the foundation for grepMo — one specific problem, one specific customer.
Stack
Django 5Django REST FrameworkReact 18TypeScriptSupabase PostgreSQLWeasyPrint

Services

Workflow automation

Connect the tools you already use so the manual steps disappear.

  • n8n workflows
  • API integrations
  • error handling and alerting
Discuss this →

Document & report automation

Turn raw data or form input into finished, formatted documents.

  • PDF/DOCX generation
  • conditional templates
  • batch output
Discuss this →

Internal tools & MVPs

A working internal product, built and deployed, not a prototype deck.

  • full-stack build
  • auth and billing
  • deployment and handover
Discuss this →

Stack

AI & AUTOMATION

  • Claude
  • GPT-4
  • Cursor
  • n8n
  • prompt engineering
  • AI-assisted development

BACKEND

  • Python
  • Django 5
  • Django REST Framework
  • REST API design
  • Supabase PostgreSQL

FRONTEND

  • React 18
  • TypeScript
  • HTML/CSS

DEPLOY & OPS

  • Railway
  • Cloudflare
  • Linux
  • Git/GitHub
  • WeasyPrint
  • Razorpay

About

I build software with AI — Claude, Cursor, and n8n — and I ship it to production. That method took grepMo from a problem I noticed to a live, paying product built entirely on my own: architecture, backend, frontend, payments, deployment, and the content that brings people to it.

B.E. Mechanical Engineering, 2017. CCNA certified. Machine Learning Specialization, DeepLearning.AI.

Based in Pune, India — overlapping US and EU hours. Available for contract, freelance, and full-time work.

Working on something?

Whether it's a product to build, a process to automate, or an internal tool you don't have time for — I'd be glad to hear about it. Replies within a day.