ERPNext
ERPNext is a full open-source ERP built on the Frappe framework, covering accounting, inventory, manufacturing, CRM, projects and HR in one integrated system. It is a serious alternative to the costly mid-market suites, with no licence fees per user and complete access to your own data and code, and it deploys cleanly with Docker.
View the repo · frappe/erpnext ↗Growing businesses that have outgrown spreadsheets and disconnected tools and want their operational and financial records in one place under their own control. It rewards firms willing to adapt some processes to the system rather than customising everything on day one. A business that only needs bookkeeping will find a dedicated accounting package simpler, and heavily bespoke operations should budget real implementation effort whoever the vendor is.
Install itgit clone https://github.com/frappe/frappe_docker
cd frappe_docker && docker compose -f pwd.yml up -dInstallation is an afternoon; implementation is the project. Chart of accounts, item and warehouse structure, opening balances and user roles all need decisions before the software helps anyone, and data migration from the old system is where timelines slip. Customise through the sanctioned Frappe extension points rather than patching core code, or every upgrade becomes a fight. Scope one module group first and expand once it is live and reconciled.
Where Blash AI comes inWe scope the modules that matter, migrate and reconcile your data, and then build what makes the ERP worth having: AI automations on top, such as document extraction with Unstructured feeding purchase invoices in, n8n workflows moving data between ERPNext and the tools around it, and Metabase dashboards over the clean database it creates.
When you want this running on your real stack, that is the engagement
Book an AI audit →n8n →
n8n is workflow automation you run on your own infrastructure. It connects the applications a business already uses through a visual editor, with hundreds of prebuilt nodes, a code step for anything bespoke, and native support for calling language models mid-flow. Think of it as a self-hosted alternative to Zapier or Make that you own outright, including every piece of data passing through it.
Ollama →
Ollama runs open-weight language models on your own machine or server. One command downloads a model and serves it behind a local API that most existing AI tooling can talk to, so an application built against a hosted provider can point at your own hardware instead. Sensitive data stays inside your network, there are no per-token fees, and it keeps working with no internet connection at all.
LiteLLM →
LiteLLM is a gateway that lets you call models from Anthropic, OpenAI, Google and many other providers through one consistent API. It runs as a Python library inside your code or as a standalone proxy server in front of your whole stack, with cost tracking, per-key budgets, rate limits and automatic fallbacks for when a provider has a bad day.