andrewyng/translation-agent

Health v2 · maintenance only

Flags: no_releases

Adoption not part of the score

5804 stars · 708 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

A Python demonstration library by Andrew Ng implementing an agentic machine translation workflow: an LLM translates text, reflects on its own output, and applies suggestions to improve it. It is explicitly a research demo rather than production software.

Use cases

• translate text with an llm using a reflection workflow
• control translation style and formality via prompts
• enforce consistent terminology with a glossary in machine translation
• translate for a specific regional dialect like latin american spanish
• generate parallel text corpora for training translation models

When to choose

• you want steerable, prompt-controlled machine translation with custom style or terminology
• you are experimenting with agentic LLM workflows for translation research
• you need to generate training data (parallel corpora) for traditional MT systems

When to avoid

• you need fast, cheap, production-grade translation at scale
• you require consistently better quality than commercial MT offerings
• you need a mature, well-supported translation product

Facets

library · maturity experimental

llm-inference agent-framework nlp prompt-engineering large-language-models python cross-platform machine-translation reflection-workflow agentic-translation llm translation natural-language-processing ai-agents localization

Member repositories

For agents

markdown · JSON · MCP: product_card(name="andrewyng/translation-agent")