Skip to content
Currently available — for the right work·France903+ Day French Streak·2026 Q2 calendar — open now
Part of ELOEmbedded Logic Operations · Full-stack lead, hands-on
↩ Projects · 2018 · procurement → offer → ledger
Still in production · support on demand · Full-stack lead, hands-on (via ELO) · Dhaka, Bangladesh

Pharmacon

Pharmacon is a turnkey ERP for one of Bangladesh's largest pharmaceutical turnkey-solution providers — the kind of business where a single deal starts with a client's plant requirement, runs through international principals for offers and quotations, continues into local buying, and ends in a ledger entry months later. I built it hands-on as the full-stack lead; it was my primary project through 2018 and 2019. The client still runs it today, and the last support round was in 2024, on demand.

Role
Full-stack lead, hands-on (via ELO)engineering
Period
2018 — 2019 · last support 2024completed
Stack
Full-stack webMulti-module ERP · Quotation & offer workflow · Ledger & dues tracking
Location
Dhaka, Bangladeshremote
VISITOFFERREVISELEDGERACKCOMPAREREVISIONHO FINALPOPROCUREMENT → LEDGER
01 · The shape of the business

The shape of the business

Turnkey plant solutions do not fit a standard sales CRM. A deal begins as a client requirement gathered on a visit, becomes a request to international principals, comes back as offers and quotations that need comparing, gets negotiated and revised, then turns into local buying, billing, and dues that outlive the original conversation by months.

Before the system, that lived in documents, inboxes, and people's memory. The ERP made each deal a tracked object with a status, a value, an owner, and a history — which is the difference between a pipeline you can report on and a pipeline you can only ask about.

02 · What the system does

What the system does

  • Client visits and requirement collection. The front of the funnel captured as structured data rather than as a note.
  • International offers and quotations. Principal, item, model, and quantity carried through the offer with cost and client-facing value side by side.
  • Status-tracked negotiation. An explicit lifecycle — acknowledged, comparison sheet, price revision, head-office finalisation, at risk, closing date, lost, purchase order — with every change user- and time-stamped.
  • Local buying and ledger. Purchase, billing, pending-bill calculation, dues and receivables tracked to the deal that created them.
  • Nudges on stagnation. Email notifications and reminders on sales workflows that stop moving, so ageing deals surface themselves.
  • Permissions and referencing. Module-level access per team, plus dynamic document IDs with stable URLs so a specific offer can be referenced in a conversation.
  • Reporting. Module-level and company-level performance in tables, charts, and graphs.
03 · Why it lasted

Why it lasted

A system a client is still running six years later, with only on-demand support, is a design outcome rather than a maintenance achievement. It survived because the model matched the business — deals, offers, and ledger entries as they actually exist in a turnkey pharma supplier — instead of forcing the business into a generic CRM shape.

It is also the project where I learned that enterprise software earns trust in the boring places: reliable notifications, permission boundaries people can predict, and reports whose numbers match what the finance team already believes.

Engineering Deep Dive

High-calibre architecture decisions.

Technical Highlights

  • End-to-end deal lifecycle: visit → requirement → international offer → negotiation → local purchase → ledger.
  • User- and time-stamped status history on every offer document.
  • Cost / offer / estimated-profit view carried on the deal record.
  • Stagnation detection with email reminders on idle sales workflows.
  • Module-scoped permissions and stable document URLs for referencing.
  • Tabular and graphical reporting at module and company level.
Performance Metrics

By the numbers.

2018 — 2019
Primary build
Procurement → ledger
One system
Still running
Last support 2024
Offer workflow
Status-tracked end to end
Collaborate

If we worked together.

I bring high-leverage product engineering and absolute operational discipline: shipping weekly, pressure-testing user inputs, and automating delivery pipelines.

Next case study

BSMMU Paediatric Nephrology

BSMMU — Bangabandhu Sheikh Mujib Medical University, now BMU, still known locally as PG Hospital — is one of the largest public hospitals in Bangladesh. I led the early build of the records platform for its paediatric nephrology department: indoor and outdoor patients, visit-by-visit entries, and disease tagging designed so the department could use its own caseload as a research corpus. It was primarily a 2018–2019 project, upgraded again in 2023–2024, and it is still actively run today. ELO is now working with them on a donations system as well.