Tacit doesn’t search your product.
It understands how it behaves.

When your users ask a question, Tacit traces how your product actually behaves — surfacing not just the answer, but every condition, exception, and rule they’d otherwise discover the hard way. Fewer tickets. Faster answers. No documentation required.

Request Early Access See How It Works
↓ Support ticket volume ↑ User activation No docs required On-prem available
The problem

Your product is more capable than your users will ever discover alone.

Not because it’s broken — because it’s deep. Complex products have conditions, exceptions, and rules that live in the logic, not in documentation. Users ask the question they can formulate. They get a partial answer. The rest stays hidden until they hit a wall, file a ticket, or give up.

The difference

Other tools answer the question.
Tacit answers the whole truth.

Most tools find the closest match to what your user typed and stop there. Tacit traces the full behavior around the question — surfacing the conditions, exceptions, and rules that govern what they’re trying to do. The answer they needed, including the parts they didn’t know to ask about.

Generic AI Tool
What happens if a transaction fails?
If a transaction fails, you’ll typically receive an error notification. Check your dashboard for failed transactions and contact support if the issue persists. Most platforms offer automatic retry options in their settings.
No match found — fell back to generic training data
Tacit
What happens if a transaction fails?
A failed transaction triggers this sequence:

1. Retried automatically — up to 3 times, with a 15-minute backoff between attempts.
2. After the third failure, status moves to Pending Review and held for up to 48 hours.
3. The Operations Manager role is notified after the first failure; Finance Managers only after the third.
4. If unresolved after 48 hours, the transaction is voided and written to the audit trail.

Note: transactions above $50,000 skip the retry flow entirely and go directly to Pending Review — requiring manual approval before any retry is attempted.
Transaction processing → Retry & backoff policy
Notification rules → Role-based alerting thresholds
High-value rules → Manual review threshold

↑ None of the branches above were asked about. Tacit surfaced them because they govern what the user is actually trying to understand.

How it works

Set up in hours, not weeks.

Tacit doesn’t search your code like a document. It maps how your product behaves — tracing connections between endpoints, services, rules, and roles — so it can answer questions that no single file contains.

Supports .NET / ASP.NET Core · Java / Spring Boot · Node.js / NestJS / Express · React / Angular / Vue

Step 01

Your team connects the codebase

A one-time setup by your engineering team — typically under a day. They point Tacit at your repository. You don’t need to be involved. You just see the result.

$ tacit index --repo ./src
Step 02

Behavior is mapped, not just indexed

Controllers, services, DTOs, routes, middleware, and business rules are traced as connected flows — not chunked as flat text. Tacit builds a map of what your product actually does.

✓ 847 endpoints · 234 services · 1,291 DTOs mapped
Step 03

Your users get a guide

A single embed in your product. Users ask questions in plain language and get complete, accurate answers — including the conditions and exceptions they didn’t know to ask about.

<script src="tacit.js" data-project="acme">
Use cases

Built for products too deep to explain themselves.

Surface what users didn’t know to ask

Users ask small questions. Tacit answers the full picture — including the conditions, exceptions, and adjacent capabilities they’d only discover by hitting a wall. Fewer support tickets as a side effect, not the goal.

Close the time-to-value gap

New users explore by asking. Tacit guides them through your product’s depth from day one — without a help center, without a walkthrough, without waiting for onboarding calls that never get booked.

Turn questions into product insight

Every unanswered question is a signal. Tacit shows you what users are trying to do but can’t find — turning every interaction into a data point that reveals where your capabilities are hidden.

Deployment & Security

On-premise and air-gapped. Your code never leaves your network.

Built for regulated industries and security-conscious teams. Tacit runs entirely inside your infrastructure — air-gapped, on-prem, or private cloud. No data ever reaches our servers.

Air-gapped deployment

Tacit ships as a self-contained package — Docker images, local LLM inference, and embedded vector search. Deploy on your own hardware with zero external network calls. No API keys to an outside service. No telemetry. No phone-home.

Designed for environments where code cannot leave the building: defense contractors, financial institutions, healthcare platforms, and any organization operating under strict data residency requirements.

Full air-gap supported
Indexing
Your codebase → Tacit → Behavior map
↓
At query time
User question → Tacit → Grounded answer
✓ Everything runs inside your perimeter. Nothing leaves.

On-prem & private cloud

Deploy on bare metal, VMware, or your private Kubernetes cluster. Supports Azure Private, AWS VPC, and any Docker-compatible environment.

Runs on your hardware — GPU or CPU-only configurations
No cloud dependency, no SaaS data pipeline
Integrates with your existing SSO and access controls

Data sovereignty & compliance

Your source code, embeddings, and user queries never leave your jurisdiction. Built for teams operating under GDPR, SOC 2, ISO 27001, and industry-specific regulations.

EU, UK, IL, and other jurisdictional deployments
Full audit trail on every query and answer
No training on your data — ever
FAQ

Frequently asked questions

How does Tacit work without documentation?
Tacit maps your codebase as connected behavior — controllers, services, routes, validation rules, and business logic traced as flows, not flat text. When users ask questions, Tacit traverses that map to surface accurate answers grounded in how your product actually works, not generic training data.
Can Tacit be deployed on-premise or in air-gapped environments?
Yes. Tacit ships as a self-contained package with Docker images, local LLM inference, and embedded vector search. It runs entirely inside your infrastructure with zero external network calls — suitable for defense, financial, and healthcare environments with strict data residency requirements.
What tech stacks does Tacit support?
.NET (ASP.NET Core), Java (Spring Boot), and Node.js (NestJS, Express) backends, along with React, Angular, and Vue frontends. Additional framework support is on the roadmap.
How does Tacit surface capabilities users didn’t know to ask about?
When a user asks a question, Tacit doesn’t just find the closest match — it traces the full behavior around what they’re trying to do. That means the answer includes conditions, exceptions, permissions, and adjacent capabilities that live in the logic around their question. Users get the complete picture, not just the happy path. Support ticket reduction follows as a natural side effect.
Is Tacit a replacement for writing documentation?
Documentation captures the happy path. Tacit traces how your product actually behaves — including every condition, exception, and rule that never makes it into a guide. Whether you have docs or not, Tacit gives users the complete picture, not just what someone had time to write down.
How is Tacit different from ChatGPT, Claude, Gemini, or Perplexity?
General-purpose LLMs — ChatGPT, Claude, Gemini, Copilot, Perplexity, Mistral, Llama — generate plausible-sounding answers from training data. They don't know your product. Tacit is grounded in your actual codebase, so every answer reflects how your product really works. It cites specific workflows and features, not guesses.
Early Access

Your users deserve the full answer.

We're onboarding a small group of B2B product teams. Get early access, direct support, and a say in what we build next.

No credit card required. We'll reach out within 48 hours.

Built by a CTO with 15+ years shipping enterprise systems across medical, construction, IoT, logistics, IPaaS, and real estate — and the last few building RAG pipelines and LLMs. Born from the same pain: features that outrun any documentation effort.