I replaced every function in a codebase with English
it still works
Details
- External ID
- 47476954
- Source
- HN
- Company
- —
- Product
- I replaced every function in a codebase with English
- Website domain
- tril.cc
- Launched
- March 22, 2026
- Cohort
- —
- Upvotes
- 9
- Upvotes percentile
- 0.5393603936039361
- Tags
- —
- Fetched at
- Sept. 7, 2026, 9:26 p.m.
- Updated at
- Sept. 7, 2026, 9:26 p.m.
Description
Tril converts every function in a repo into plain English descriptions, then runs and tests them — using an LLM as the interpreter instead of a runtime.The concept: programming languages exist because machines couldn't understand human intent. LLMs can. So what happens if you remove the code entirely and just... describe what each function should do?The tool replaces functions one by one, runs your test suite after each swap to confirm nothing breaks, and outputs a .md file. Then `tril run` spins up an HTTP server that sends each function's English description to Claude and returns the result.Tested on a unit converter (JS) and a 625-line Python CLI tool — tests passed, results matched to 6 decimal places (fortunately).This is mainly a thought experiment: can any code become plain natural language? Will it still work? Let's find out!npm: npx @sliday/tril convert URLGitHub: https://github.com/sliday/tril
Enrichment
- Theme
- ai translation and localization tools
- Vertical
- Horizontal
- Function
- Dev tools
- Audience
- Developer
- AI stance
- Not AI
- Project type
- Hobby / open-source project
- Normalized one-liner
- code documentation replacement experiment
- Manually corrected
- False
Could you build this?
Yes The project is a developer experiment that reads ASTs/code files, prompts an LLM to generate plain English descriptions, and executes test suites by prompting an LLM to evaluate results.
Discussion
6 comments analyzed.
Competitors mentioned: COBOL (English-like programming language), Yggdrasil (intent-to-code project)
Concerns raised: LLM output non-deterministic and unreproducible based on intent, Token cost prohibitively high compared to code output size, Security and stability of AI-generated code questionable, Premise flawed - replacing working code with English-to-LLM-to-code adds unnecessary steps, Loop creates black box without intermediate precision guarantees
Competitors
Other products that read as similar to this one — 109 launches clear the similarity bar, closest 8 shown.
Attention rank: #60 of 110 (itself plus its competitors, highest first — normalized so YC and Product Hunt are compared fairly).
Launched 141 days after the earliest competitor.
- An application stack Claude coded directly in LLVM IR · hn · 2026-03-12 · 10 upvotes · similarity 0.42
- OtterLang · hn · 2025-11-08 · 15 upvotes · similarity 0.41
- The Tsonic Programming Language · hn · 2026-01-13 · 62 upvotes · similarity 0.41
- Trilyo · ph · 2026-09-17 · 2 upvotes · similarity 0.40
- Openleetcode · hn · 2026-08-11 · 5 upvotes · similarity 0.39
- Openleetcode · hn · 2026-08-18 · 73 upvotes · similarity 0.39
- Genotype: language that compiles to idiomatic TS, Rust and Python · hn · 2026-09-08 · 9 upvotes · similarity 0.38
- Fuse · hn · 2026-08-02 · 121 upvotes · similarity 0.38
Other launches for this product
- No other launches for this product.
Same idea, different domain
Nobody's really built a dev tools tool for Sales yet.