AI-Native Translation Platform

AI agents that translate,
review, and QA for you.

Autonomous agents run entire jobs. LLMs work as translation engines. AI chat sees every segment. You stay in control — step in anytime.

TilBridge TMS product-guide.xliff EN → DE
Translator agent completed · 118 of 127 segments translated Done
# Source Target
1 Getting Started
Erste Schritte 101 TM
2 Click {1}Save{/1} to confirm.
Klicken Sie auf {1}Speichern{/1}. Agent
3 Your changes have been saved successfully.
Ihre Änderungen wurden erfolgreich gespeichert.| Agent
4 Maximum file size: 50 MB
Maximale Dateigröße: 50 MB Agent
5 Click here to continue.
Klicken Sie hier, um fortzufahren. Agent
Confirmed: 118 / 127
93%
1,284 words · 7,821 chars
14
Agent tools
<100ms
TM lookup
45+
File formats
6+
LLM providers

Capabilities

AI-native from day one

AI Agents

Autonomous agents that translate, review, and QA entire jobs. Assign them to workflow steps like any team member.

Agent Translator
Pre-translate → Review → QA
14 tools Token budget: 50K

AI Chat

A conversational assistant that sees every segment, TM match, and term base in your project. Full context, instant answers.

LLM Translation

Use OpenAI, Gemini, or self-hosted models as translation engines. Tag-safe output with cost tracking.

AI Match Repair

When a TM match is close but not perfect, AI adapts the translation. Word-level diffs show exactly what changed.

RAG Pipeline

Upload style guides, briefs, previous translations. Documents are chunked, embedded, and searchable — AI always has context.

Translation Memory

Exact, fuzzy, and semantic matching. Word-level diffs on every match. Sub-100ms lookups on 1M+ entries.

Rich Text Editor

ProseMirror-powered with inline tag handling, WYSIWYG and All Tags modes. Full RTL and CJK support.

Quality Assurance

14+ automated checks gate on confirmation. Tag consistency, numbers, terminology, spelling — errors caught before they ship.

45+ File Formats

XLIFF 1.2/2.0, vendor bilinguals, plus native document formats. Round-trip reconstruction preserves every style and formula.

Other CAT tools bolt AI onto

decades-old architecture.

We built AI into

every layer from day one.

AI Agents — Translator Agent
Running
Translator
3-phase post-edit
Briefing

Post-edit MT output. Preserve tone, fix terminology, ensure tag safety. Use TM for consistency.

Tools 9 of 14 enabled
read_segmentsquery_tmquery_tbget_mtpre_translatewrite_targetsrun_qa_jobconfirm_segmentsadd_comment
Safety
Tokens 50K
Iterations 30
Batch 20
Phase 2 — Reviewing segments
iter 14/30
18.2K tokens 42 segments 8.3s elapsed

AI Agents

Autonomous AI that translates for you

Create AI agents that translate, review, and QA entire jobs on their own. Assign them to workflow steps like any team member — they run in the background and report when done.

Visual agent builder

Drag-and-drop canvas to configure behavior, tools, and safety limits. No code required.

Pre-built templates

Start with Translator (3-phase post-edit), QA Reviewer, or build your own from scratch.

14 specialized tools

Agents read segments, query TM and term bases, run QA checks, write translations, and search the web.

Safety & observability

Token budgets, iteration caps, dry-run mode, and real-time action logs. Cancel anytime.

AI Chat — product-manual.xliff

Is #12 consistent with #5? They both mention the same feature.

AI

Good catch. #12 translates "Translation Memory" as "Übersetzungsspeicher", but #5 uses "Translation-Speicher". Your term base has "Übersetzungsspeicher" as the approved term — I'd update #5.

Ask about your translation…
342 in · 89 out Session: 4 messages

AI Assistant

Ask your AI about the entire job

A conversational assistant that sees every segment, translation memory, and term base in your project. Ask about terminology consistency, source ambiguity, or anything else — it has the full context.

Full job context

The AI sees every segment, TM match, and term base entry — no setup needed.

Segment references

Mention segments by number. The AI links back to the exact ones it discusses.

Instant answers

Responses stream in real time. No waiting for the full reply.

Per-job memory

Each job has its own conversation that picks up where you left off.

Machine Translation — Segment 8
Source (EN)

Press the key to export your {1}document{/1}.

DeepL Popular

Drücken Sie die Taste, um Ihr Dokument zu exportieren.

ai:openai AI

Drücken Sie die Taste, um Ihr {1}Dokument{/1} zu exportieren.

ai:anthropic AI

Betätigen Sie die Taste, um Ihr Dokument zu exportieren.

3 engines · 1.2s
All tags preserved

AI Translation

Any LLM as a translation engine

Plug in OpenAI, Gemini, or self-hosted models alongside traditional MT providers. Same interface, same tag safety, same caching — just a different brain behind the translation.

Tag-safe output

Inline tags are preserved exactly. Formatting stays intact after translation.

Any language pair

LLMs handle virtually any language. No per-pair configuration needed.

Multiple providers

Use OpenAI, Gemini, or self-hosted models — all in the same interface.

Cost tracking

Token usage is tracked per engine, per job. Responses are cached to save costs.

Match Repair

AI fixes what fuzzy matching can't

When a TM match is close but not perfect, the AI adapts the translation to reflect source changes. Minimum edits, maximum fidelity — the original translation is treated as gold standard.

Automatic repair

Triggers automatically for the first fuzzy match. Toggle AI on or off anytime.

Word-level diff

See exactly what the AI changed — additions, deletions, and modifications highlighted.

One-click apply

Accept the repaired version with a single click, or ignore it and use the original.

Preserves everything

Tone, terminology, style, and inline tags are kept intact. Only the diff is adapted.

TM Match — Segment 23
AI
TM Match 87%
Source diff

Click the buttonPress the key to export your reportdocument.

TGT

Klicken Sie auf die Schaltfläche, um Ihren Bericht zu exportieren.

TGT-AI

Klicken Sie auf die SchaltflächeDrücken Sie die Taste, um Ihren BerichtDokument zu exportieren.

2 changes 0.3s

Intelligence

MT, AI, and semantic search

Multiple translation engines, AI-powered improvements, and search by meaning. Choose your providers, control your costs.

DeepL

Popular

High quality for European languages with XML tag preservation

Google Cloud

Broadest language coverage — 100+ languages

OpenAI

AI

Context-aware translation with custom prompts

Regional providers

Local

Tilmoch, MangiSoz, and other locally built engines from Central Asia and CIS regions

Self-hosted

OpenAI-compatible API for your own models

Key capabilities

Tag-safe translation
Response caching
Cost tracking per engine
Configurable fallback threshold
Batch pre-translation

Reference Materials

Give your AI the full picture

Upload style guides, client briefs, previous translations, terminology sheets, or even URLs. Documents are chunked, embedded, and searchable — so AI agents and the editor chat always have the right context.

Any document type

PDFs, DOCX, XLSX terminology sheets, URLs, previous translations, annotated documents — all supported out of the box.

Automatic chunking & embedding

Documents are split into sections, embedded, and indexed automatically. No manual tagging or configuration needed.

Previous translation alignment

Upload source and translated documents as a pair. TilBridge aligns them automatically — even when paragraphs don't match one-to-one.

AI-powered retrieval

Agents and chat search references by meaning, not keywords. Relevant excerpts surface automatically per segment in the CAT panel.

References — Legal Translation Project
4 documents
PDF

Client_Style_Guide_DE.pdf

Style Guide
DE 48 chunks
Indexed
DOCX

Previous_Contract_EN-DE.docx

Prev. Translation
EN, DE 124 pairs aligned
Indexed

client.example.com/legal-glossary

URL
EN 12 chunks
Indexed
XLSX

GDPR_Terminology_Multilingual.xlsx

Terminology
EN, DE, FR, RU 340 terms
Indexed
search_references("formal register for contracts")
0.94 Style Guide › Chapter 3 › Register Rules

"All contractual obligations must use the formal register. Use Vertrag not Deal. Avoid colloquial expressions in legal contexts."

0.89 Prev. Translation › Clause 4.2

"The controller shall implement appropriate measures." → "Der Verantwortliche setzt geeignete Maßnahmen um."

0.82 GDPR Terminology › Row 47

EN: data controller → DE: Verantwortlicher — Legal term, GDPR Art. 4(7)

Context summary AI-generated

Covers formality requirements for German contract language. Specifies formal register, approved legal terminology, and tone guidelines from the client's style guide.

Hybrid retrieval dense + sparse
524 chunks indexed
12ms

The Editor

Step in anytime — it's still your editor

AI handles the heavy lifting, but you're always in control. A professional CAT editor built from scratch — faster, more efficient, and modern.

Term enforcement

Approved terms underlined in source, forbidden terms struck through

QA on confirm

Checks run automatically — errors block confirmation until fixed or overridden

Number & tag checks

Catches mismatched numbers, missing tags, and broken formatting

15+ check types

Terminology, consistency, whitespace, URLs, brackets, doubled words, and more

TilBridge — user-manual.xliff
QA issues found — press Ctrl+Enter again to override
# Source Target
5 Welcome to TilBridge TMS. Willkommen bei TilBridge TMS.
6 Update the database settings. Datenbankeinstellungen aktualisieren.
7 Maximum file size: 50 MB
Maximale Dateigröße: 500 MB
8 erase the old records
die alten Einträge löschen
QA Issues 1 error 1 warning
E

#7 · Numbers

Number "50" in source but "500" in target

W

#8 · Terminology

Forbidden term "löschen" — use "entfernen"

Jobs project-manual.xliff — Editor
EN → DE | 247 segments | 3,842 words
B
I
U
1 User Guide
Benutzerhandbuch 101 TM
2 Getting Started
Erste Schritte 100 TM
3 Click {1}Save{/1} to confirm your changes.
Klicken Sie auf {1}Speichern{/1}, um Ihre Änderungen zu bestätigen. 95 TM
4 The maximum file size is 50 MB. Die maximale Dateigröße beträgt 50 MB.|
5 Export your project as PDF or DOCX.
6 Contact support for assistance.
7 Version 2.1 Release Notes

Translation Memory

Three ways to find the right match

Most tools give you exact and fuzzy. TilBridge adds semantic search — finding translations by meaning, not just surface similarity.

Translation Memory Panel
Current Segment

Save your changes

Exact Match 100%
Match

Save your changes

Translation

Speichern Sie Ihre Änderungen

Hash Trigram Embed Auto
100%

Exact Match

Finds identical source text instantly. In-context matches (101%) verify surrounding context for extra confidence.

87%

Fuzzy Match

Finds similar segments even when wording differs. Word-level diff highlighting shows exactly what changed between the match and your segment.

0.94

Semantic Match

Multilingual embeddings find semantically equivalent translations despite completely different wording. Matches competitors miss.

Auto

Smart Propagation

Confirm a segment once and identical segments fill automatically. Pre-translation applies matches on import — 101% auto-confirmed, lower matches saved as drafts.

QA Results — project-manual.xliff
Segment 47 1 error
Source

Use the Translation Memory to check.

Target

Verwenden Sie den Translation-Speicher zur Überprüfung.

Terminology: approved term is "Übersetzungsspeicher", not "Translation-Speicher"
Checks Run
Tag consistency
Number match
Terminology
Trailing punct.
Double spaces
Untranslated
Bracket balance
URL preserved
Ctrl+Enter to confirm Blocked — fix errors first

Quality Assurance

Catch errors before they ship

QA runs automatically when you confirm a segment. Errors block confirmation — you fix the issue or consciously override. No silent mistakes.

On-the-fly gates

Errors block confirmation. Override with a second Ctrl+Enter if intentional.

Full-job analysis

Run QA across all segments. Results persist and link to each segment.

Terminology enforcement

Approved terms are required. Forbidden terms are flagged.

Locale-aware

"1,000.50" (EN) correctly matches "1.000,50" (DE). No false positives.

Search & Navigation

Find anything, instantly

Global search, concordance, find & replace with regex, and powerful segment filters — all with live highlighting.

Cmd+K opens a unified search across segments, TM, term bases, and semantic matches. Live results as you type, keyboard-navigable, with scope tabs and type badges.

certificate of origin | 4 results
All Segments TM TB Semantic
Certificate of Origin Ursprungszeugnis
The certificate of origin must be signed Das Ursprungszeugnis muss unterschrieben werden
Attach the original certificate of origin Fügen Sie das Original-Ursprungszeugnis bei
Document proving product origin Dokument zum Nachweis der Produktherkunft

Workflow & Collaboration

More than just an editor

Revision History

Every edit tracked with origin — manual, TM match, MT, propagation, AI. Smart dedup keeps the audit trail clean without keystroke noise.

Analysis & Pre-Translation

Run TM analysis to see leverage across 8 match bands before work begins. Then pre-translate — auto-fill from TM, MT, or both with configurable thresholds.

Workflow Steps

Translation → Review pipeline with role-based permissions. Each step tracks who edited what. Advance, rollback, and complete with full audit.

Comments & Discussion

Per-segment threaded comments. Discuss translations with your team. Rich text support with user attribution and timestamps.

Batch Operations

Filter segments, then act on all at once — confirm, lock, unlock, copy source to target. One-click bulk actions without selecting individually.

Term Base

Glossary with approved and forbidden terms. Auto-suggest popup while translating. Instant lookup even with 10,000+ entries.

Project Management

Assign agents or people — or both

Chat with the project, ask AI to draft a brief, assign agents to workflow steps alongside human translators. Every job, resource, and agent — on one page.

Projects / Acme Corp Q1 / user-manual-v3.xliff EN → DE
Open Editor
Download
Workflow
Translation
Translator Agent
Review
Michael R.
Done
Advance
Rollback
Step 2 of 2 · Started Jan 15
File Overview
Segments
247
Repetitions
18 (6 groups)
Words
12,450
Characters
74,820
Analysis
Run Analysis
Leverage
67%
Weighted Words
4,109
101%
22%
100%
15%
Fuzzy
30%
New
33%
Translation Memories
Attach TM
Main TM
R/W Penalty: 0
Client Ref TM
Read Penalty: 2
Reference Materials
Upload
Style Guide v2.pdf
524 chunks
Previous EN→DE.xlsx
Aligned
247 segments · 12,450 words Last analysis: 2 hours ago

Why TilBridge

What makes it different

Legacy CAT Tools
TilBridge TMS
Architecture
Java/.NET desktop, 2000s codebase
Built from scratch with modern tech
Performance
Seconds for fuzzy on large TMs
Instant results, even on huge memories
Grid Scroll
Laggy above 5K segments
Smooth scrolling with 10,000+ segments
AI Agents
Not available
Autonomous translation, QA, and review agents
Reference Materials
Shared folder, no search
Upload style guides and briefs — AI searches them automatically
TM Lookup
Exact + fuzzy only
Exact, fuzzy, and meaning-based search
Tag Handling
Basic tag protection
WYSIWYG + All Tags modes, tag color cycling
Source Diff
None or basic
Word-level diff highlighting on every match
QA
Run manually, review a list
Gates on confirmation, blocks errors
File Formats
15-20 via plugins
45+ formats out of the box
Collaboration
File locking, no shared TMs
Simultaneous TM access across jobs and users
Desktop App
Electron or native (heavy)
Lightweight desktop app, works offline

Technology

Built on a modern foundation

Built for speed and reliability — launches fast, runs smoothly, and never gets in your way.

Rust Backend

Handles millions of translation memory entries without breaking a sweat.

axumtantivysqlxtokio

Svelte 5 Frontend

A responsive, snappy interface that feels instant — no loading spinners.

svelte 5prosemirrortailwind v4vite

Desktop via Tauri 2

Work offline with a lightweight desktop app. No heavy installs. Coming soon.

tauri 2webviewofflinenative

Free Trial

Try TilBridge TMS for Free

We're onboarding new users. Leave your details and we'll set up your workspace.