Tackle complex processes
We build AI into the tangled, multi-step processes inside an organization — the work that's too specific or too messy for off-the-shelf software to handle.
AI solutions company
PotionLabs builds AI into the complex, costly, day-to-day processes most software ignores — making them faster and leaner, and surfacing the problems you couldn't see. Our first product is an AI-native ERP for manufacturers.
● Proven inside JD Jones & Co. — a ~100-year-old fluid-sealing manufacturer that exports 85–90% of what it makes — and now built into products for businesses like it.
what we do
PotionLabs is an AI solutions company. We put AI to work inside the processes that run a business — wherever it makes the work simpler, faster, cheaper, or clearer.
We build AI into the tangled, multi-step processes inside an organization — the work that's too specific or too messy for off-the-shelf software to handle.
AI that makes business processes faster and more efficient — collapsing work that took hours into minutes, without adding headcount.
AI solutions that cut operational and process cost — removing the manual, repetitive effort that quietly eats into margin.
AI that resolves problems you couldn't see — surfacing what's buried in your data and daily workflows before it turns expensive.
the problem
Most businesses already have software. The real work still runs on WhatsApp.
Those systems were built for the finance department — the ledger, the audit trail, the month-end close. The operational work was an afterthought. So the people who actually run it quietly went back to spreadsheets and a group chat.
And the chat works — messages get answered in minutes, everyone sees everything, the history is searchable. The real competition for operations software was never SAP. It's the group chat. So we build AI that's faster than the chat, for the people who actually do the work.
our first product
We proved the approach the hard way — inside a real, demanding factory. The result is a complete AI-native ERP that runs a manufacturer's operational core — orders, production, quality, procurement and dispatch — with the AI layer built in. It's the first product on the PotionLabs platform.
Explore the AI-native ERP →what it runs
The books stay in Tally or SAP.
how we work
We start from the documents and workflows you already have — not a multi-quarter data migration. Your system of record doesn't change on day one; we just remove the manual middle, then build out from there.
We start from what you already have — your PO PDFs, invoices, packing lists, Tally exports and the way work actually moves through your floor. We don't migrate you; we read you.
The platform speaks your document formats, approval chains and field names — because they were learned from your paperwork, not forced into someone else's data model. Training collapses from weeks to hours.
Once the process is encoded, AI takes the busywork — reading documents, drafting replies, surfacing the next action — while your people make every call that ships product or spends money.
the proving ground
Built and hardened inside a real, demanding factory before it became a product — which is exactly what makes it hard to copy.
Years of operation behind its proving ground
Manual data entry removed across order & invoice work
Time to turn a batch of customer POs into validated sales orders
Export share of the proving-ground manufacturer's output
where this is going
The ERP industry assumes go-live is measured in quarters. We think it should be measured in hours: walk into a factory on a Monday morning, and have orders, production and dispatch running on the system by Monday evening. Over time, own the financial ledger too. This is the ambition we're building toward — not a description of what ships today.
questions
The short, honest version of what PotionLabs is, what it does today, and what's still ahead.
PotionLabs is an AI company; its first product is an AI-native ERP that runs an export manufacturer's operational core — orders, production, quality, procurement, inventory and dispatch — with a built-in AI layer. It was proven first inside JD Jones & Co., a ~100-year-old fluid-sealing manufacturer, and is now being productized for export-focused SMB manufacturers.
An AI-native ERP is built operations-first; SAP and Tally are built around the finance department and treat operations as an afterthought. PotionLabs is shaped around how a factory actually works, with AI reading documents and drafting busywork, while the books stay in Tally or the customer's existing SAP.
No — not today. PotionLabs runs operations and leaves your financial system of record — general ledger, banking, statutory GST and e-invoicing — in Tally or SAP. A greenfield SMB runs PotionLabs instead of a heavy SAP rollout while keeping Tally for the books.
Live AI reads incoming purchase-order PDFs and extracts every line item, drafts customer replies and outreach, and answers operational questions through a conversational assistant. The rule is simple: AI proposes, humans approve anything that leaves the company or commits money. A batch of orders that took 8 hours of manual entry now takes about 2 minutes.
Going live takes weeks for a new factory today. The goal PotionLabs is building toward is a single-day operational takeover — walk in on a Monday morning and have orders, production and dispatch running by Monday evening, with the books staying in Tally. That one-day, any-size takeover is the ambition, not a claim about today.
Export-focused small and mid-sized manufacturers — businesses with real customs and trade-documentation pain (RoDTEP, SCOMET, LUT, EPCG, drawback), idiosyncratic purchase-order formats, multi-stage approvals and batch-controlled production. These are exactly the buyers SAP, Oracle and the global document-AI leaders have priced out or under-served.
Because the differentiator is depth, not breadth. PotionLabs was built and proven inside JD Jones & Co., a ~100-year-old, ~₹40 Cr fluid-sealing manufacturer that exports 85–90% of its output. Every module — from export declarations to OEE production tracking — was shaped against a demanding, real operation rather than a generic data model, which is what makes it hard for horizontal tools to copy.
PotionLabs covers sales-order and purchase-order automation, procurement with approvals and signed POs, production tracking with live OEE, quality and non-conformance registers, export documentation, and a conversational operations assistant. Financials, banking and statutory filing stay in Tally or SAP.
Yes. Human-in-the-loop is a design rule, not a setting: AI drafts and proposes, but a person approves anything that ships product, commits money or leaves the company — every time. Purchase orders pass two independent human gates (content approval and director signature) before they go out. The goal is to accelerate the operator, not remove accountability.
Pricing is implementation-led and scoped per customer, anchored well below the total cost of ownership of a SAP Business One rollout, with the AI layer included. The simplest next step is a short discovery conversation — contact PotionLabs and we will outline an approach.
A short walkthrough of where AI could take the busywork out of your operations — starting with how your orders, production and dispatch would run on PotionLabs. No procurement gauntlet, no twelve-month timeline.