Consulting
AI consulting for organisations that need operational clarity
Praestantis helps organisations evaluate AI opportunities, design workable processes, and make implementation choices that stand up in practice.
Practical AI from strategy to product
Praestantis helps organisations apply AI in ways that are operationally real from consulting and education to Clementine.
Praestantis combines AI consulting, public-facing education, and product delivery. Clementine is the flagship Android product direction: an AI receptionist for real PSTN calls with multilingual intake, operator handoff, and auditable reporting.
- AI consulting grounded in implementation
- YouTube content focused on practical AI adoption
- Clementine for real PSTN call handling on Android
Positioning
Praestantis bridges advisory work, public AI education, and product execution.
The proposition is not abstract AI commentary. Praestantis helps organisations understand AI, make deliberate choices, and turn those choices into concrete working systems such as Clementine.
Consulting
AI strategy, implementation guidance, and workflow design
YouTube
Practical videos on AI, operations, and public-sector application
Product
Clementine for multilingual PSTN call answering and operator continuity
Praestantis works where AI needs to become concrete, explainable, and useful:
AI consulting and governance
YouTube education and thought leadership
Municipal and regulated workflows
Product delivery through Clementine
What Praestantis offers
Two visible tracks: AI advisory work and the Clementine product direction.
The site needs to represent Praestantis as more than a single product. Clementine is one concrete outcome, while consulting and YouTube remain part of how Praestantis helps organisations adopt AI deliberately.
YouTube
Public AI education through practical long-form content
The YouTube track explains AI developments, workflow impact, and implementation thinking in a way that supports real decision-making.
Product
Clementine as a tangible AI product for front-office operations
Clementine shows how Praestantis turns AI ideas into a working Android product for real PSTN call answering, routing, and reporting.
Clementine
The Android product direction focuses on real phone answering, not on a demo chatbot.
Clementine is intended to become the first answering layer for real incoming calls. It can determine whether a call should stay with AI handling, be routed back to the operator, or continue with transcript capture enabled.
Answer
Handles real incoming PSTN calls on Android
Clementine is designed around Android's default phone-app role so the product direction is real phone answering, not separate VoIP infrastructure and not a call simulator.
Understand
Captures caller intent, structured intake, and multilingual speech
The app maintains live transcripts, an intake snapshot, and call notes while using local speech components and a local language model on the device.
Escalate
Returns calls to the operator with or without transcript capture
Known contacts and expected topics can be routed back into the normal phone process, while the operator can choose whether the rest of the call should continue with transcript capture enabled.
Capabilities
The product is designed around call operations, not around a chatbot demo.
The current capability set is focused on answering operations, auditability, routing logic, and practical follow-up. Every feature exists to reduce missed context and improve operator control.
01
Real PSTN answering
Uses Android phone-role integration for real carrier-call handling instead of separate VoIP support.
02
Whitelist and blacklist routing
Known contacts and expected topics can be kept in the normal phone process while unknown callers are triaged by AI.
03
Operator handoff with context
When a human needs to take over, the operator can accept the call directly or continue with transcript capture enabled.
04
Five supported languages
English, Dutch, Spanish, French, and German are built into the user-facing language strategy and speech stack.
05
Transparent logs and reporting
Each call can record the caller number, known-contact match, audit trail, transcript, call note, and daily summary email.
Typical workflow
A call can move from ringing to resolution without losing accountability.
The flow below reflects the intended operational behavior of the product rather than an abstract capability list.
Step 1
Incoming call arrives
Clementine identifies the number, checks whether the caller is known, and decides whether the call stays with AI or returns to the normal phone process.
The first decision point is operational: who called, what routing mode applies, and whether auto-answer is appropriate.
Step 2
AI triages and documents
During the conversation, Clementine can maintain a transcript, structured intake, and a continuously improving call note.
This makes it possible to hand off a live call or follow up later without reconstructing what the caller actually needed.
Step 3
Operator receives the right context
When a human should take over, the operator can accept the call normally or accept it with transcript capture enabled and receive the outcome afterwards by email.
The point is not just automation. It is reducing lost context when a call has to move between AI and a human operator.
Safeguards and control
Clementine is built to make its decisions visible.
The product direction is intentionally traceable. Operators need to know which number called, whether the caller was known, what route was chosen, which actions were attempted, and why.
Per-call transparency
- Caller number and resolved label
- Known contact match
- Transport and routing decision
- Action audit with reasons
Local-first handling
- On-device language model
- Local speech recognition and TTS
- Background model download at first start
- Reduced dependency on remote AI services
Practical reporting
- Transcript email at call end when enabled
- Automatic daily call log email
- No daily message when no calls occurred
- Encrypted storage for mailbox credentials
Availability
Clementine is a Praestantis product direction for serious call handling.
The current build direction is Android-first, multilingual, and operationally traceable. It is suitable for organisations that want a controlled AI receptionist rather than a black-box AI phone bot.
Contact
Talk to Praestantis about consulting, content, or Clementine.
If you want to discuss AI consulting, YouTube collaboration, multilingual call handling, operator workflows, or how Clementine fits your organisation, Praestantis can provide a focused conversation.
Leeuwte 18
7963 PK Ruinen