# UI Options
allow_amplitudepermite selectare gate pentru tipologie de extindere input (QMS, QMQ)
allow_inptafișează inpt
allow_alphaafișează inpt alpha
allow_gammaafișează inpt gamma
allow_betaafișează inpt beta
allow_modelpermite selectare model pentru creator
allow_imagespermite adaugarea de fisiere
allow_stylepermite alegerea stilului
is_translatoractiveaza selectoare de limba SURA - TARGET (translator Univers ALL)
allow_dimensionspermite selectare text dimensiune implicita
allow_formattingpermite selectare format implicit
allow_inputspermite TrainsComprehension Launch Pad
allow_transhraphPermite alegerea transgraphului
---title: RENDA ADN---
# RENDA - Platform
# index
- [RENDA - Platform](#renda---platform)- [index](#index)- [definitions](#definitions)- [CONCEPTS](#concepts) - [GATE](#gate) - [Reorganizing strategies:](#reorganizing-strategies)- [How to read this?](#how-to-read-this)- [Concepte și tehnologii utilizate](#concepte-și-tehnologii-utilizate)- [JOURNEY](#journey)- [FAIRO Search](#fairo-search)- [Interactions](#interactions)- [TransLingua (TLANG)](#translingua-tlang)- [T-LANG query language:](#t-lang-query-language)- [Tech Stack Description](#tech-stack-description) - [**Backend Layer**](#backend-layer) - [**Frontend Layer**](#frontend-layer)- [MONOVOCE EAR](#monovoce-ear)- [MONOVOCE LAB](#monovoce-lab)- [MULTIVOICE](#multivoice)- [RFC Authentication](#rfc-authentication)- [VISION LAB](#vision-lab)- [MEMORY](#memory)
# definitions
In the rapidly evolving field of artificial intelligence (AI), it is essential to understand the various roles, components, and frameworks that contribute to the development, implementation, and management of AI systems. This introduction aims to provide an overview of six key concepts in the AI domain: mechAInisms, WorkAIflows, FrameAIworks, trAIners, facilitAItors, and navigAItors.
These six concepts provide a foundation for understanding the complex landscape of AI roles, components, and frameworks, enabling professionals and organizations to navigate and capitalize on the transformative power of artificial intelligence.
setAining - este procesul prin care dezvoltatorii transformă environmentul ca să fie programați de acesta în atingerea succesului.
mechAInisms - mechAInisms are advanced data analysis and processing systems within the RENDA platform, specially designed to obtain predictable and relevant results on various data sets. trAIners play a crucial role in refining and instructing the models and algorithms involved in mechAInisms, utilizing their expertise in different domains and analytical techniques to improve the performance of AI systems. mechAInisms dynamically adapt to various types of data and domains, providing a robust and scalable framework for analyzing and interpreting data to optimize decisions and actions within an organization.
chAIn- WorkAIflows are essential components of the RENDA platform, offering a comprehensive set of tools and resources to support the development and deployment of artificial intelligence workflows, focusing on integrating multiple mechAInisms to create a consolidated result. facilitAItors and navigAItors collaborate within the WorkAIflows component to ensure the efficient use of AI technologies in work processes and in achieving organizational objectives. They manage and optimize AI workflows by combining different mechAInisms for data analysis and processing, extracting valuable information and generating efficient solutions, while promoting adaptability and innovation in AI projects.
FrameAIworks - FrameAIworks, integral parts of the RENDA platform, are the most powerful sets of tools designed to integrate automation bots, advanced logic, adaptive tools, mechAInisms, and WorkAIflows, with the aim of creating products that surpass the level of expertise, intelligence, and human creativity in specialized domains. integrAItors apply their transdisciplinary knowledge to connect and blend various fields with artificial intelligence, facilitating cooperation between different areas of specialization and improving the application of AI in a coherent and harmonious manner. These comprehensive frameworks support AI developers, facilitAItors, and integrAItors in creating innovative solutions that transcend the limits of traditional human knowledge while ensuring long-term compatibility and scalability to meet the challenges of specialized domains.
trAIners - trAIners are specialists who take on the role of instructing and refining artificial intelligence models and algorithms within mechAInisms, as well as training facilitAItors and navigAItors to effectively utilize WorkAIflows and FrameAIworks. With extensive experience in various domains and analytical techniques, they use their knowledge to analyze, interpret, and improve the performance of AI systems. Through interdisciplinary collaboration, trAIners contribute to the creation of innovative solutions and technological advancements in the field of artificial intelligence, while ensuring that facilitAItors and navigAItors are well-equipped to manage and optimize AI components within an organization.
facilitAItors - facilitAItors are experts in managing and implementing artificial intelligence systems within an organization, utilizing WorkAIflows and FrameAIworks for the efficient integration of mechAInisms. They possess interdisciplinary skills and adaptability in various working environments and collaborate with trAIners, navigAItors, and diverse teams to ensure the seamless integration and efficient use of AI technologies in work processes and in achieving organizational objectives. facilitAItors play a central role in promoting the adoption of artificial intelligence and stimulating innovation within companies.
navigAItors - navigAItors are professionals who, through WorkAIflows platforms, manage and optimize AI workflows within an organization, ensuring the effective implementation of mechAInisms and FrameAIworks. Although they do not always require advanced technical training, navigAItors possess interdisciplinary skills and a solid understanding of how AI systems integrate into various domains, working closely with trAIners and facilitAItors to develop their expertise. They ensure the efficient and effective use of AI technologies and collaborate with diverse teams to improve processes and outcomes.
# CONCEPTS
## GATE
EGO: "ego", KN_EGO: "knego", MECHAINISM: "mechainism",
LOGBOOK: "logbook",
// DINAMIC STEPLESS WORKAIFLOW SELF_TXT: "self_txt",
WORKAIFLOW: "workaiflow",
PORTAL: "portal",
FN_CALL: "fn",
PERSON: "",
// SPECIAL RESEARCH TYPE RESEARCH: "research",
GEN_IMAGE: "generative_image", GEN_AUDIO: "generative_audio", GEN_VIDEO: "generative_video", GEN_CODEX: "generative_codex",
// used as a redirect MODULE: "module"
Reorganizing strategies:-
a. direct
b. dpr (decompose - process - recompose) configurable - full parallel - tabula scripta - previous
c. self.txt - autopilot - step by step
Applied tactics:- markandseep
# How to read this?
This is a technical specification on how RENDA platform is implemented.Renda is composed of components called GATES which act as INPUT->PROCESS->OUTPUT
# Concepte și tehnologii utilizate
în sistemul de Automatizare Maximală a livrabilelor de proiectare, părți scrise.
sess[A]Ion:
sesiune de lucru, atașată de operațiunile întreprinse în RENDA de un utilizator.
gate: reprezintă un sitem complex, de tip blackbox inteligent ce permite procesări de tip: `IN -> GATE { processing } -> OUT`
workaiFlow: Un [AI]lgoritm (suită de pași) implementat cu gate-uri. WorkAiFlows sunt de două tipuri: - autonome - asistate (similar cu SELF.TXT). Doar anumite workaiflows sunt de tip asistat pentru navigators. Creator poate in orice workaiflow sa intre pe modul asistat. Poți opri la fiecare pas și poți edita inputurile fiecărui pas.
Tr[AI]nsLingua: Limbajul utilizat în Platforam Renda pentru a comunica cu infosystemul și a determina facerile din cadrul sistemului.
storage:
1. storage_reference - Reprezintă ID_ul unic din RENDA acordat unui spațiu de stocare. Acesta este utilizat pentru automatizare prin standardizare.
2. logbook_vego: Spațiu de stocare (de tip TransGraph) pentru surse în legătură cu un proiect sau un al subiect asupra căruia se pot genera livrabile. (e.g. proiect, o persoană în cazul unui sistem HR - putem considera fiecare „persoană un proiect în sine”, ca stocare)
3. common_knowledge: Spațiu de stocare (de tip TransGraph) pentru documente de Tier înalt, aplicabile într-o gamă largă de proiecte: legislație, cărți, etc. ce sunt utile în general
scenar[AI]o: Reprezintă un sistem de „currying” (matematic -o funcție \( f(x, y) \), generează o funcție parțială \( g(x) \) prin fixarea valorii lui \( y \)).
Scenariile sunt fixate și legate de livrabile standard. Outputul aplicării unui scenariu este un livrabil standard. Un scenariu apeleaza in spate GATES sau WorkAiFlows cu legătură de „storage_reference”.
Journ[AI]:
Singularitățile: Singularitățile se genereaza direct din TransGraph, pentru toate tipologiile de documente pe care le definim că suportă singularități. Acestea sunt exportate apoi într-un livrabil standard numit singularități și se autoactualizează pe măsură ce se adaugă alte documente. Va exista un BOT care va detecta dacă apar conflicte și va trimite notificare la șeful de proeict să rezolve conflictele.
# JOURNEY
Journey is the ultimate automation. Is achieved through a set of features:
1. SESSION 2. Scenario 3. WorkAiFlow 4. Subject 5. continuum
Activate a session to track usage. Singularitățile se genereaza direct din TransGraph, pentru toate tipologiile de documente pe care le definim că suportă singularități. Acestea sunt exportate apoi într-un livrabil standard numit singularități și se autoactualizează pe măsură ce se adaugă alte documente. Va exista un BOT care va detecta dacă apar conflicte și va trimite notificare la șeful de proeict să rezolve conflictele.
# FAIRO Search
Tabula Omnia Building.
Each FILE has a DEFAULT formatting.Each SEARCH CONFIGURATION has an override OPTION which allows complete
# Interactions
O interacțiune este o „procesare„ / „cerere„ / „mesaj„ transmis în RENDA.
# TransLingua (TLANG)
TLANG is a contextual communication language used in RENDA Ecosystem.
there are two kinds of variables:- references - used for advanced data retrieval- direct vars - used for storing intermediary values
modifiers:? - optional modifier When used this means that the whole segment$ - variable@ - reference@$ - extract variable value@ - indicates that a reference is being constructed@: - references the corresponding bucket
When inserting references, the following syntax is used:
After insertion references are resolved to the corresponding values> @:: - references the corresponding key
or if the key is an action the corresponding action is applied
> @:.. - references the corresponding key and applies the action
```t-lang
$_settings = {{ $variable = sfasfs $label = waf DENUMIRE CE II APARE UTILIZATORULUI $description =
$prompt_type = 'STANDARD' | 'BETA' | 'GAMMA' | 'DELTA' $model = defined $ultimate_power = yes | no $sef-focus = yes | no $expert = $amplius_mode = }}
|STEP> $_settings = {{ $label = s1_response $gate = 44883f72-162a-400d-8d6c-2762a4f2cf69 WAF 5 WHY $parsing_pattern = !!!! $parsin_max_block_size = 60000 $processing = dpr | dpr_up_down | epr $model = default $model_temperature = 0 $model_max_response_tokens = 4000 $model_top_p = 0 }}
$inpt = {{ $label }}
$_settings = {{ $label = insert number here $gate = 44883f72-162a-400d-8d6c-2762a4f2cf69 WAF 5 WHY $parsing_pattern = !!!! $parsin_max_block_size = 60000 $processing = 7 $model = $model_temperature = 0 $model_max_response_tokens = 4000 $model_top_p = 0 }}
$inpt = {{ @$inpt }}
$rist =
$ctxt = {{ Aici avem context din @$s1 si @$s0 }}
$tmsp =
$inpt =
$tgrf =
: - references the corresponding bucket
When inserting references, the following syntax is used:
After insertion references are resolved to the corresponding values- @:: - references the corresponding key
or if the key is an action the corresponding action is applied
- @:.> - references the corresponding key and applies the action
ACTIONS: direct search: N-1, N+1, N search
action chaining: ..
# Tech Stack Description
The application is constructed using a modern and efficient technology stack aimed at delivering high performance, scalability, and an interactive user experience. The stack is composed of robust data storage systems, a high-speed backend runtime, and lightweight frontend libraries.
---
**Data Storage Layer**
**PostgreSQL**
- **Role**: Serves as the primary relational database management system (RDBMS) for storing persistent application data.- **Features**: - Advanced querying capabilities. - Support for complex transactions. - Strong data integrity and reliability.- **Usage**: Ideal for handling structured data and ensuring ACID (Atomicity, Consistency, Isolation, Durability) properties in database transactions.
**Redis**
- **Role**: Utilized for caching, session management, and as a message broker.- **Features**: - In-memory data storage for fast data access. - Supports data structures like strings, hashes, lists, sets, and sorted sets. - Pub/Sub messaging capabilities.- **Usage**: Enhances application performance by reducing database load, managing user sessions efficiently, and facilitating real-time communication.
---
## **Backend Layer**
**JavaScript (Runtime: Bun.js)**
- **Role**: Powers the server-side logic of the application.- **Bun.js Features**: - High-performance JavaScript runtime and package manager. - Faster startup and execution times compared to traditional Node.js environments. - Built-in support for TypeScript and JSX.- **Usage**: Offers improved performance and developer experience, accelerating server-side operations and enabling efficient handling of client requests.
---
## **Frontend Layer**
**HTML5, JavaScript, CSS**
- **Role**: Forms the foundation of the user interface, ensuring structure, styling, and interactivity.- **Features**: - **HTML5**: Semantic elements and multimedia support. - **CSS**: Responsive design and visual styling. - **JavaScript**: Client-side scripting for dynamic content.
**Kita.js**
- **Role**: A JavaScript library or framework used to structure the frontend application logic.- **Features**: - May provide components or utilities to enhance development efficiency. - Could offer architectural patterns to streamline frontend development.- **Usage**: Complements other frontend technologies to build a responsive and interactive user interface.
**HTMX**
- **Role**: Enables dynamic web applications by extending HTML with advanced features.- **Features**: - Allows HTML attributes to trigger HTTP requests. - Supports AJAX, CSS transitions, WebSockets, and Server-Sent Events directly in HTML.- **Usage**: Simplifies the creation of interactive web pages without relying on heavy JavaScript frameworks, facilitating partial page updates and improving user experience.
**Alpine.js**
- **Role**: Provides a minimalist approach to adding interactivity to web pages.- **Features**: - Declarative and reactive components. - Small footprint with easy integration. - Similar syntax to larger frameworks like Vue.js.- **Usage**: Ideal for handling simple interactions and state management, enhancing the frontend without significant overhead.
# MONOVOCE EAR
**Quick Features:**- **Audio Whisper**: Quick, efficient audio processing.
**Two Modes:**- **Two Live Options**: - **Conciliere Renda Transcribe**: Emulate live audio transcription with two-line support. [Link to Renda Transcribe](https://beta.renda.holdings/chat?uuid=3e193513-5960-4eb6-a65d-d3f8089006b3)
**Enhanced Version:**- **EAR Renda Monovoce Enhanced**: Improved live transcription and audio emulation. [Link to Enhanced Version](https://beta.renda.holdings/chat?uuid=13dd6960-262d-40c3-9b14-8993f0ddafdf)
---
# MONOVOCE LAB
**Quick Features:**- **Audio Whisper**: Quick and streamlined audio whispering.
**Emulate Live Options:**- **Variant A**: Renda Audio Single Line Monovoce [Variant A Link](https://beta.renda.holdings/chat?uuid=audio_v1)
- **Variant B**: DPR version for audio emulation [Variant B Link](https://beta.renda.holdings/gates/audio_v2)
**Enhanced Version:**- **EAR Renda Monovoce Enhanced**: Improved functionality for both variants. [Enhanced Link](https://beta.renda.holdings/chat?uuid=13dd6960-262d-40c3-9b14-8993f0ddafdf)
# MULTIVOICE
- **Azure**: Multi-speaker support with high quality.- **Audio Flow**: - **First Line**: Audio Whisper - **Second Line**: Renda Audio Single Line Monovoce Variant A [Second Line Link](https://beta.renda.holdings/chat?uuid=audio_v1)
**Conciliator Multivoce:**- **EAR Renda Multivoce Enhanced**: Multi-speaker enhanced version for seamless audio. [Conciliator Link](https://beta.renda.holdings/gates/636a781a-ba9a-4452-9179-7b3b5478cea9)
# RFC Authentication
Authentication - cookie based.Supported auth mechanisms:- [X] user & password- [X] microsfot auth- [ ] google auth
Permissions & Right Management.
Role based auth: - guest - navigAItor, - facilitAItor, - integrAItor, - trAIner - creAItor
Fleets - users are grouped into fleets.
# VISION LAB
Before you uploadEvaluate image sizeYou can include multiple images in a single request (up to 5 for claude.ai and 100 for API requests). Claude will analyze all provided images when formulating its response. This can be helpful for comparing or contrasting images.
For optimal performance, we recommend resizing images before uploading if they exceed size or token limits. If your image’s long edge is more than 1568 pixels, or your image is more than ~1,600 tokens, it will first be scaled down, preserving aspect ratio, until it’s within the size limits.
If your input image is too large and needs to be resized, it will increase latency of time-to-first-token, without giving you any additional model performance. Very small images under 200 pixels on any given edge may degrade performance.
To improve time-to-first-token, we recommend resizing images to no more than 1.15 megapixels (and within 1568 pixels in both dimensions).Here is a table of maximum image sizes accepted by our API that will not be resized for common aspect ratios. With the Claude 3.5 Sonnet model, these images use approximately 1,600 tokens and around $4.80/1K images.
Aspect ratio Image size1:1 1092x1092 px3:4 951x1268 px2:3 896x1344 px9:16 819x1456 px1:2 784x1568 px
Calculate image costsEach image you include in a request to Claude counts towards your token usage. To calculate the approximate cost, multiply the approximate number of image tokens by the per-token price of the model you’re using.
If your image does not need to be resized, you can estimate the number of tokens used through this algorithm: tokens = (width px * height px)/750
Here are examples of approximate tokenization and costs for different image sizes within our API’s size constraints based on Claude 3.5 Sonnet per-token price of $3 per million input tokens:
Image size # of Tokens Cost / image Cost / 1K images200x200 px(0.04 megapixels) ~54 ~$0.00016 ~$0.161000x1000 px(1 megapixel) ~1334 ~$0.004 ~$4.001092x1092 px(1.19 megapixels) ~1590 ~$0.0048 ~$4.80Ensuring image qualityWhen providing images to Claude, keep the following in mind for best results:
Image format: Use a supported image format: JPEG, PNG, GIF, or WebP.Image clarity: Ensure images are clear and not too blurry or pixelated.Text: If the image contains important text, make sure it’s legible and not too small. Avoid cropping out key visual context just to enlarge the text.
Activare Memorie Human Memory de la data - ora cutare: yyyymmddHHMMss last_M_interactionsActivare Memorie Tabula Scripta de la data - ora cutare: yyyymmddHHMMss last_M_interactionsActivare memorie Ideologică de la data - ora cutare: yyyymmddHHMMss last_M_interactions only on this/ only on the project / on all gates
tip memorie se selecteaza din select box si ai 3: Human | Tabula | Ideologica
data_de la care sa inceapa: last_X_time (minutes)
d:yyyymmddHHMMss | i:10
Fhairo pe Proiect
tii fhairo pe proiect dar doar pe gate-ul curent
On ALl
# MEMORY
MERGE CognitionMutare Istoric + Scenarii Superclara in TB Super Clara
