{# Analytics is consent-gated: the Google script is not requested at all, and no cookie can be set, until the visitor accepts. #}

What I build

Software engineering at Crispa, and projects of my own in stochastic modelling, software, and the occasional circuit.


Valuation Insights
Engineer, full stack
valuation-insights.crispa.ai

Built and shipped a production SaaS turning Denmark's public company register into valuation, ownership and market data, working as an engineer across the stack.

Backend
Django / DRF + Postgres
20 apps, ~160k LOC
Frontend
Next.js / TypeScript
~64k LOC
  • Designed the LLM pipeline that parses free-text Danish legal filings into validated capital-history chains: every raise, reduction, merger and demerger, cross-checked against structured register data.
  • Built the valuation engine, an ownership graph resolving ultimate control across holding structures, a market-comparables matrix, and a feed poller maintaining millions of rows.
  • Shipped the full commercial layer: Auth0, Stripe billing with a webhook-synced mirror, transactional email, Sentry-correlated audit logging over 1.2M rows, and a HubSpot lead-gen integration.
  • Owned infrastructure end to end: Dockerised multi-service deploy on Fly.io (backend, frontend, Postgres, Redis), CI/CD and scheduled GitHub Actions pipelines running the data-refresh jobs and standing health checks that alert on silent pipeline degradation, with Sentry observability across the stack.
  • Optimised production Postgres with partial indexes and query-plan analysis, cutting multi-second scans to sub-millisecond.


@teodoras66
Source for these projects lives on GitHub. Follow along or dig into the code.

If you'd like to talk about any of this, email me or find me on LinkedIn.