My Avatar

Junaid Anjum

Product Engineer based in Srinagar, Kashmir.

Viscos

Overview

Process safety and risk assessment at Bell Energy.

Role

Product interfaces, the Atlas worksheet engine, and frontend integration with study and risk assessment APIs.

Stack

TypeScript, React, Next.js, TanStack Query, Zod.

Problem

A hazard study connects equipment, deviations, causes, consequences, and safeguards. Engineers need to edit that information as a worksheet while preserving the relationships behind each row.

My work

I build the study interface, reusable template and risk libraries, and workspaces with role-based access. These connect the organisation of a facility to the studies engineers carry out within it.

For AI-assisted work, I built a review interface where engineers inspect suggested study entries and accept or reject them before they enter the worksheet. I also build the interfaces connecting hazard studies to protection-layer assessments and safety function verification.

Atlas

Making a tree editable as a table

I built Atlas to turn hierarchical study data into a spreadsheet interface. Parent cells span their related rows, keeping context visible as the worksheet grows.

The engine supports keyboard navigation, cell editing, copy and paste, and undo/redo. Its data operations are separate from React rendering, so changes to the hierarchy and changes to its presentation have distinct responsibilities.

Recent work

I added ordered background saving and recovery for worksheet edits, alongside reusable worksheet views that new studies inherit from their template.

I built a visual model canvas for safety functions, where engineers arrange sensors, logic solvers, and final elements, select the layouts used in calculation, and inspect results alongside the design.

The interface distinguishes previous and outdated results after edits, while keeping historical runs and their input evidence available for review.

Result

Engineers can edit related study information in context, reuse configured risk data, and review AI suggestions explicitly. The contribution described here is the product interface and grid engine; the backend services are developed with the wider team.