I led the technical development aspect of a web-based economic analysis tool for evaluating project viability in a climate/industrial domain. The primary users were engineers and economists who defined projects with detailed technical and financial parameters and then computed a levelized unit-cost metric used to assess break-even economics under discounted cash-flow assumptions.

I built the application foundation: a Django backend with data models, an extensible input form system, interactive Plotly/Dash visualizations, integration with an enterprise data platform, and deployment into an existing Kubernetes environment. The core calculations engine (discounted cash flow, NPV/IRR, levelized-cost solving) was implemented by domain specialists. I designed a clean serialization boundary between the application layer and the calculation layer so each could evolve independently without leaking internal details.

The input workflow was the most complex UI work. Scenario definitions included dozens of parameters across multiple categories (timelines, ramp-up profiles, consumables/utilities, labor, transport/storage, and detailed capital and operating cost line items). I implemented dynamic Django forms with inline formsets for variable-length structures, strong validation, and a draft/error workflow: if a calculation fails, the scenario is saved as a draft with an actionable error message so the user can correct inputs and retry safely.

The output workflow consisted of three interactive views:
– A summary view with KPI tiles and breakdown charts across major cost components.
– A detailed view with an annualized table spanning the full project lifetime.
– A sensitivity analysis view where users select variables, define ranges, and visualize their impact on the levelized-cost metric.

Each save produced a versioned snapshot in the data platform to support auditability and traceability. The service was operated on Kubernetes with rolling deployments, monitoring/alerting, availability objectives, and SSO-based authentication.

The team was small (up to three developers). I served as technical lead and primary backend/infra engineer, partnering closely with a domain specialist for the financial model and an additional developer focused on UI. The tool was delivered to production and used to evaluate real investment scenarios.

Client name and identifying details withheld due to confidentiality. Screens and data are recreated/illustrative; no proprietary assets included.