Codcel turns mission-critical spreadsheets — insurance pricing, actuarial models, risk calculations, quant models — into production-grade Rust, Java, or Python services your IT team owns, audits, and deploys.
Benchmarked on real pricing, reserving, and risk models — millions of rows, hundreds of rating factors, full audit trails.

That's not a tooling problem. It's an architecture problem. Codcel fixes it.
Your model in Excel

Generated code, in seconds
use std::error::Error;
#[derive(Debug, Clone)]
pub struct QuoteInput {
pub driver_age: i32,
pub postcode: String,
pub vehicle_value: f64,
pub no_claims_years: i32,
}
#[derive(Debug, Clone)]
pub struct QuoteOutput {
pub annual_premium: f64,
pub monthly_premium: f64,
pub base_rate: f64,
pub no_claims_discount: f64,
}
fn main() -> Result<(), Box<dyn Error + Send + Sync>> {
let input = QuoteInput {
driver_age: 35,
postcode: "SW1A 1AA".to_string(),
vehicle_value: 24_500.0,
no_claims_years: 5,
};
let output = codcel_quote(&input)?;
println!("annual_premium: {}", output.annual_premium);
println!("monthly_premium: {}", output.monthly_premium);
println!("base_rate: {}", output.base_rate);
println!("no_claims_discount: {}", output.no_claims_discount);
Ok(())
}Every factor table, every lookup, every formula — translated into type-safe code your IT team reviews, tests against your current outputs, and deploys behind your existing API gateway.
For developers| Internal rewrite | Low-code platform | Specialist vendor | Keep Excel | Codcel | |
|---|---|---|---|---|---|
| Time to production | 12–24 months | 3–6 months | 6–12 months | Instant | Days |
| Business team keeps Excel | No | No | No | Yes | Yes |
| Source code you own | Yes | No | No | N/A | Yes |
| Runs on your infra | Yes | Sometimes | No | N/A | Yes |
| Audit-ready governance | Yes | Partial | Yes | No | Yes |
| Handles 10M+ rows | Depends | No | Yes | No | Yes |
Codcel is the only option where business teams keep Excel and IT gets real, owned, production code with full governance.
Turn rating engines, factor tables, and reserving models into governed services.
See insuranceTurn capital models, stress tests, and IRB calculations into audit-ready services.
See bankingTurn trading models, pricing curves, and scenario engines into production-grade APIs.
See quantAlso used for energy trading, tax engines, healthcare actuarial, and other regulated calculation workloads.
Millions of rows
Benchmarked on real pricing, reserving, and risk datasets — not toy examples.
Hundreds of factors
Nested lookups, tiered factors, and multi-dimensional rating tables stay correct under load.
Sub-millisecond per calc
Compiled Rust services keep quote and pricing latency well below request budgets.
Model in Excel.
Domain experts tag inputs and outputs with small inline markers. Existing formulas untouched.
Generate and review.
Codcel emits idiomatic code your engineers review, test, and commit to Git.
Deploy like any service.
Your existing CI/CD, observability, and security posture apply. No new runtime, no new vendor.
Codcel generates real, production-ready source code across today's most widely used programming languages and ecosystems — with more coming soon.
Rust
Lib · API · WASM · Web
Java
Lib · API · Web
Kotlin
Lib · API · Web (incl. Android)
.NET (C# / VB)
Lib · API · Web
Python
Lib · Web
Go
Lib · API · Web
JavaScript
Web Clients
TypeScript
Web Clients
Swift
Lib · Apple
FFI / C
Native Bindings
AI · MCP
Agent Tools
WASM
WebAssembly Modules
We're continuously expanding Codcel's language and platform support.
From a single Excel file, Codcel produces full systems — APIs, UIs, libraries, databases, and even MCP servers for AI agents. No templates, no frameworks, no lock-in.
A fully documented REST API built from your Excel logic. Deterministic, type-safe, and ready to deploy to any cloud or on-prem environment.
Generate idiomatic libraries in Rust, Python, Java, C#, Go, Swift, TypeScript and more — readable, testable, and production-ready.
Turn your Excel tables into SQL schemas and high-speed and CSV import for massive datasets (including 1B+ rows).
A complete React-based UI with forms, validation, tables, and live calculation logic connected to your generated backend.
Expose your Excel logic as AI tools. Let ChatGPT or your internal LLM safely call calculations, run scenarios, generate reports, and quote premiums.
Export your Excel logic as a fast, portable WebAssembly module — perfect for embedding directly into browsers, desktop clients, or edge environments.
All generated artifacts are real, human-readable source code — fully owned by you, versioned in your repos, deployed in your architecture, and governed by your processes.
We're partnering with a small number of insurers, banks, and financial institutions to turn their Excel calculation engines into governed, auditable services.
Design partners get deep engineering support, preferential pricing, and direct influence on the product roadmap.
Talk to us about your book
For insurers, banks and actuarial teams turning production models into governed, auditable services. We run the demo on your file, and design partners get deep engineering support and preferential pricing.
Try it yourself
For developers and small teams who want to see real generated code from a real Excel file, no call required. Start on the Explore plan from €15/month, full source, all languages, cancel anytime.
Demos take 30 minutes. Bring your hardest spreadsheet.