Decide.

TaranDM®.
 Founded in 2019
Southeast Asia (HQ in Singapore), Europe (Prague), North America (Toronto)
Services and solutions in decision management, big data architecture & advanced data analytics

Our
footprint
Key Differentiators

PaaS, not SaaS Stateful, end-to-end decisioning

TaranDM runs on client’s infrastructure Fast and easy data integrations to other internal & external
(cloud/on-premise) systems
We don’t store or have access to any Data pre-processing (ETL)
production data. Production data don’t leave Seamless deployment of AI/ML models
client’s infrastructure Decision flow management & deployment via web-based GUI
(including data source calls)
Out-of-the-box simulation capabilities
Decisioning dashboards
No data sharing with the vendor.
No IT security, compliance or regulatory
concerns. Business benefits (data
readily available for post-analysis, …). Business users in control. Less dependence
Lower costs. on IT/engineering. Unrivalled flexibility &
time-to-market.
Automate and improve complex, data-driven decisions
across customer’s lifecycle
Use case versatility
Both real-time & batch decisioning

ORIGINATION CUSTOMER MANAGEMENT COLLECTIONS

Customer service
Revenue & profit optimization ma
management
ment
Portfolio management

Churn strategiess
Pers
Personalized offers
ns
Loan applications Recovery strategies
Cro
Cross-sell & up-sell Retention strategiess
n
Offer personalization
Early phase
Pre-collection strategiess strategies
Fraud detection
La phase
Late
strategies
p
Mid phase
Pre-qualification
ificati offers strategies
TaranDM design & technology stack
– minimizing the vendor lock-in

Microservice IT security Flexible deployment Open-source technologies High performance

architecture SSO (Single Sing-On) options & easy & standards & scalability
integrations No proprietary languages or formats
Field-level encryption
TaranDM Deployment
Taran infrastructure Client infrastructure (cloud/on-premise)
Internet (HTTPS/VPN)

TaranDM Core backend

(Taran AWS)

Docker (container) images Production environment (Kubernetes)

Configuration
TaranDM Develop environment Testing environment (Kubernetes)
for each client
(Taran AWS Kubernetes)
Docker (container) images Other environments – UAT, etc. (if needed/requested)

Develop environment for the client is available from day 1 Production data don’t leave client’s infrastructure
of the project TaranDM production run doesn’t rely on vendor’s
All development testing done at Taran infrastructure, which leads infrastructure and its availability
to significant reduction in development time for new client-specific
functionalities and features, such as new data adapters, etc. If required, fully hosted deployment possible as well
Traditional, process-driven decisioning
Reject

Loan origination Decision engine Loan origination Decision engine Loan origination Decision engine
system PRE scoring system POST scoring system FINAL scoring

Approve
Internal data External data Request income proof,
collection Reject collection Reject run manual underwriting,
etc.

Data-driven decisioning
Reject
Initial data Approve
or additional data
Loan origination
Decision engine
system
Request more data or action:
income proof, manual Gather data
underwriting, etc. (internal, external)

Advantages
One decisioning strategy for the full approval process
Configuration done in GUI, less dependence on IT/engineering teams
Flexible data collection with possible time and cost savings
More convenient post-decision analysis
Faster & simpler data integrations, configuration and testing
Traditional, deterministic strategy assignment

Credit card application

Decision engine
(Credit card strategy)

Loan origination system

Decision engine
(Cash loan strategy)
Determines which decisioning Cash loan application
strategy to call

Dynamic strategy assignment

Credit card application
Credit card strategy
Application data
Credit info (Cash loan vs. Credit card)
Loan origination
Decision engine
system

Decision engine assigns Cash loan strategy

strategy automatically Cash loan application
based on credit type

Advantages
Flexible to add new credit products, reusing current strategies, no need for IT changes
Easier testing and maintenance
Traditional deployment architecture

Develop environment Testing environment Production environment

Manual transfer Manual transfer

All environments completely separated

Changes done in non-production environments have to be transferred manually to upstream environments
Non-production environments usually contain outdated strategies, which makes testing less effective and time-consuming
Client charged for each environment

“Smart synchronization” deployment architecture

Smart synchronization of strategies & business configuration across environments

Develop environment Testing environment Production environment

Sandbox Simulation Live Sandbox Simulation Live Sandbox Simulation Live

All environments separated from operational and data point of view. Smart synchronization mechanism for strategies & business configuration
Changes done in downstream environments are automatically available in upstream environments’ sandboxes as well
All downstream environments can be synchronized with the production environment via single click in the GUI
Client can have as many environments as requested for no additional charge
No limitation on number of users and “work-in-progress” projects. Users can work and prepare the changes concurrently, share with others, etc.
TaranDM Functional Diagram

state:
FINISHED

APPROVED
REJECTED
CANCELLED
External / Internal
data source

Client’s infrastructure 1st API call with TaranDM state:

(triggering system) input data (Decisioning Core) GET_DATA ...

External / Internal
data source
state:
ACTIONS_NEEDED

Subsequest API call(s)

with additional data
TaranDM Architecture
Internal & External Data Sources

TaranDM GUI AntiFraud

Cache DB

Data Sources Adapters

Simulation DB
Controller
Simulation
Core DB

Decisioning Core
Embedded Analytical
Real-time Batch Environment
module module

Reporting

Frontend
Appendix
TaranDM
Client Stories
Client Story

Project objectives
Modernizing the existing decisioning infrastructure
Consumer finance company with a portfolio and automating the approval process
of tailored financial products Having the ability to set up product parameters
and configuration easily
Subsidiary of Abdul Latif Jameel, a Saudi
Empowering the business users to do & deploy
Arabian conglomerate with more than
75 years history the changes without the need of IT

[Link]
Why has ALJ Finance Egypt
chosen TaranDM
The team – Taran has experienced team which
covers expertise from credit risk, antifraud,
modelling and scorecard areas. We do not
usually see such well-rounded support during
pre- and post-launch, not only from technical
infrastructure and interface perspectives, but
Taran team also accommodated our inquiries
from the business strategy areas
The product – TaranDM has refined technology
“Greatly appreciate stack and offered features which are highly
the swift response and relevant in addressing pain points from credit
underwriting capabilities perspectives
hands-on support provided
by Taran team, as well as Implementation highlights
their flexibility and readiness Automation of the underwriting processes which

to proactively step in have been mostly manual in the old system

Integration with several data source and ability
during periods where to consume them in the real time

we had internal resource The Egyptian Credit Bureau (iscore)

Behavioral data
capacity challenges.” Internal blacklists
Onsite training to ensure the smooth knowledge
Nicole Oesman transfer to client’s team. The business users are
Sector Chief Risk Officer now self-sufficient and are able to modify data
input & output, create new business logic or
change the existing one, test it and deploy to
production by themselves
Client Story

Neo Financial’s
adjudication goals
Replace the existing credit decisioning infrastructure with
an enterprise decision engine which allows a high level
Neo Financial of flexibility to credit risk managers in designing
and testing multiple adjudication strategies and
processes simultaneously
One of the largest, fastest growing, and most
Allow for faster time-to-market when deploying new
popular digital banks in Canada decisioning logic or its changes
Eliminate dependency on engineering and developer
Valuation > USD 1bn (unicorn status) teams and give credit risk managers the ability to have
Website: [Link] end-to-end ownership of their processes

Why has Neo Financial

chosen TaranDM
“Platform-as-a-service” way of delivery – the decision
“As a fast-growing fintech, engine runs at the client’s infrastructure (either cloud
or on-prem), no sensitive data is shared with
we need tools and partners the vendor (customers’ personal details, etc.)
Streamlined implementation & process migration,

that are going to support us measured in weeks, and not in many months or years
Modern technology stack – containerized, Python-based

in our growth while allowing decision engine with microservices architecture

Multiple out-of-the-box functionalities usually not

us to move at a fast pace. available in other decision engines, such as Impact

Report (simulation of decisioning logic’s changes

Taran has enabled us to against the status quo), batch decisioning, or Decision
Optimizer

implement a wholistic
Implementation highlights
credit adjudication strategy
TaranDM implemented and Neo Financial’s existing credit
to effectively gain new adjudication process migrated in less than 3 months
Continuous knowledge transfer during the implementation
customers and give our & phase-in, along with TaranDM’s general customizability,
has ensured that Neo Financial’s internal teams will not
team the tools to manage be overly dependent on us as the vendor going forward
Higher flexibility when deploying new decisioning logic,
our existing portfolio.” for example:
Credit product pricing now determined by
the strategy and client historical profile/credit
Brent Bishop, Chief Credit Risk Officer
bureau record
Specific strategies for selected partners which allow
fine-tuning of approval rates/risk profiles and
support of marketing actions and rollout plans
Seamless A/B testing of strategies for various client
segments
Client Update

Neo Financial
One of the largest, fastest growing, and
most popular digital banks in Canada

Valuation > USD 1bn (unicorn status)

Website: [Link]

Neo Financial has been using TaranDM for more than

a year now. What has been its business impact?
“We are now leveraging more data to enhance our decision-making process. With TaranDM,
we can utilize 20× more data points in the adjudication process, and adding new ones
has become incredibly easy and efficient. Previously, incorporating new data sources
required several months, but now it's much quicker.

Our credit risk management has also seen significant advancements. We’ve transitioned
from a broad adjudication path to having multiple, tailored paths for various segments.
This allows us to be more targeted and adapt to unique situations with greater precision.
Additionally, our ability to implement changes has greatly improved.

What once took weeks of discussion and planning can now be achieved within a day.
We prepare, simulate, and deploy new experiments rapidly, collecting actual data and
obtaining answers almost immediately. This has allowed us to iterate at an impressive
speed.”

Christian Rosati, Director of Risk Analytics and Credit Decisioning at Neo Financial
Client Story

Migration from a legacy

First “neobank” to secure a digital banking
license in Philippines decision engine
Operating since early 2021, focus on retail Tonik Bank had been using a decision engine
banking & consumer lending from a traditional vendor for several years and
decided to change it for a modern “cloud born”
[Link]
solution fully hosted within the bank infra-
structure

Why has Tonik Bank chosen

TaranDM
Taran offers a tool enabling simple, efficient
and flexible control of the approval processes,
including seamless integration of external data
sources
Taran showed very pro-client based approach
but was also very clear, realistic and honest in
“Tonik has acquired a powerful tool terms of implementation timeline, and did not
make any unrealizable promises
to manage its lending business,
simplifying and accelerating the
Implementation highlights
improvements we are implementing.
The implementation of TaranDM & full production
The Taran implementation team migration from the legacy infrastructure took
demonstrated a high level of a few months, including the launch of 2 credit
products (1 configured by Taran team, 1 configu-
competency and a strong focus red by Tonik Bank team themselves in TaranDM’s
on client needs. This tool is a key GUI)
Data orchestration (workflow management) has
enabler for Tonik's business
been migrated from existing Loan Origination
growth.” System to TaranDM
Tonik Bank team can now see & control their
Marek Matas end-to-end decisioning flow (all stages,
Chief Risk Officer
including data source calls) in TaranDM GUI
Client Story
Migration from a legacy
decision engine
RinggitPlus had been using a decision engine from a large
Fintech in Southeast Asia vendor for several years
The solution wasn’t flexible enough, it was technologically
Focus on retail credit products out-of-date & required support from the vendor
As a consequence, it had been significantly hindering
RinggitPlus business growth. Moreover, it was very expensive

Why has RinggitPlus chosen

TaranDM
Tremendous flexibility for the business users & risk managers.
No dependence on IT/engineering
Modern technology stack. Decisioning logic versioned,
archived, timestamped, attached to specific user & can be
easily deployed or roll-backed
IT security & compliance aspects. TaranDM runs on client’s
infrastructure (cloud/on-premise). Vendor doesn’t store or
have access to any production data
Great references from the existing clients
“TaranDM engine does
Migration scope
exactly what we wanted 2 strategies (credit card, personal loan)
from it, and some more. 3 stages (pre-bureau, post-bureau, offer optimization)
Close to 50 policy rules
It has allowed us to 5 scorecards
8 decision trees/decision tables
smoothly replace
Implementation approach
the legacy, outdated & highlights
decision engine, and The RinggitPlus was concerned that the migration will take as
kickstart our business long as the initial setup of the legacy engine (> 6 months)
However, the design of decisioning strategies in TaranDM is
growth.” significantly more efficient & flexible than in traditional decision
engines. For example:
Long decision trees are replaced with decision tables, which
Jack Low
are easier to prepare & maintain
Chief Commercial Officer
Scorecards can be uploaded in one click, no need to
manually configure binning and transformations
Decisioning logic shared by multiple strategies can be easily
re-used via so-called components. No copy pasting or
manual re-creation is needed
Ultimately, the implementation has been successfully finished
within ca 2 months, from the very start of the project to
migrating 100% of the traffic through TaranDM
Client Story
Business challenge
& objectives
For its newly launched online loan application
process, the client wanted to enhance its appro-
Ceska sporitelna val process with alternative data capturing the
(Erste Group) digital footprint of each applicant
Those “digital data” can be used for both
Founded: 1825
antifraud and scoring purposes
The largest banking group in Central and
The existing decisioning solution couldn’t process
Eastern Europe
efficiently the sizeable digital data in the real
Operations in 12 countries in the region time; a novel approach was needed
Total assets: EUR 327bn (June 2022)
Number of employees: 44,773 (June 2022)
Taran approach
Website: [Link]
Taran has designed end-to-end real-time solution
for collecting and processing of digital data from
online credit application form
Such data include device data, location data,
copy pasting activities, field changing history,
mouse movements, device fingerprint and many
“Taran has been one others

of the very few providers TaranDM is used for ca 25,000 of loan applicati-
ons every day
who have delivered their
solution under Technical details on
the TaranDM deployment
the originally agreed Designed architecture includes real-time data
timelines” collection through Kafka and NiFi, storing data
into the Elasticsearch for real-time purposes, and
Project manager at Erste Group data lake for analytical purposes
TaranDM executes attributes in Python and
calculates scores which are handed over to
classical approval process
To enable smooth scorecard deployments, Taran
has also set up Spark analytical layer with Jupyter
frontend and connected it with Git (definition of
predictors)
CI/CD done using internal Gitlab solution,
deployment on OpenShift
Client Story
Why has Silk Bank chosen
TaranDM
Silk Bank chose TaranDM for its role in modernizing credit
decisions and launching digital lending products. TaranDM's
intelligent features ensure compliance and enhance financial

Silk Bank precision. Beyond its impact on the approval process, TaranDM's
ability to seamlessly integrate with various data sources positions
it as a versatile solution for Silk Bank, ensuring a holistic and
financial institution, serving both retail informed approach to decision-making. The platform will
significantly improve our approval process by providing tailored
and corporate clients in Georgia product offers, backed by thorough credit bureau analysis.
Its seamless integration with data sources, such as CreditBureau
Website: [Link] and the Revenue Service, underscores TaranDM as a comprehensive
solution for Silk Bank. The platform's adaptability and robust
features make it an instrumental asset in Silk Bank's journey
towards modernized and efficient financial operations.

Silk Bank’s objectives

In choosing TaranDM for our bank, we are leveraging a proven
solution that has consistently delivered beyond expectations for
reputable institutions. The client stories from TaranDM previous
Launching digital lending products with fully automated projects underscore the ability to not only meet but exceed strict
approval process timelines and facilitate the seamless migration of credit
Modernizing the bank’s credit risk decisioning process assessment processes. What distinguishes TaranDM is its adaptable
and infrastructure deployment, showcased in prior project instances where it facilitated
real-time data collection and operated seamlessly in a platform-as-a-
Ability to launch any new configuration of decision-making -service framework. This adaptability aligns with Silk Bank's need for
components in days a cutting-edge solution that integrates seamlessly with diverse data
sources, empowering us to make informed decisions in real time.
Additionally, TaranDM's commitment to knowledge transfer and
customization ensures Silk Bank's teams are self-sufficient and not
overly dependent on external vendors, which is a crucial aspect
for long-term operational success. Additionally, TaranDM offers
a user-friendly interface and a no-code graphical platform,
empowering Silk Bank's team to adapt quickly to changing market
demands and configure strategies independently. Its real-time
processing capabilities ensure a swift and responsive customer
experience, aligning seamlessly with Silk Bank's commitment to
efficiency and customer satisfaction.
In essence, TaranDM is not just a tool, it’s a strategic ally in our
pursuit of excellence, combining cutting-edge data science with
a user-friendly interface to drive innovation, efficiency, and
superior customer satisfaction in Silk Bank's financial operations.
“We are working with multiple This partnership positions Silk Bank ahead in the banking industry,
empowering us to navigate the evolving landscape with agility and
vendors during the digital stay ahead in delivering unparalleled financial services.

transformation effort and TaranDM

team has been outstanding Project highlights
in delivering the complex solution. Implementation of the sophisticated income estimation logic
to be compliant with the regulatory/internal requirements
The team is responsive, timely, Enhancement of the overall approval process:
and very professional, delivery TaranDM generates all product offers for which
the customer is eligible
on time!” That includes a refinancing offer, if relevant based on
the credit bureau information
Successful and continuous know-how transition to the Silk Bank’s
Aleksi Khoroshvili, General Director team. Well before the production launch, they configured 2nd loan
product by themselves in TaranDM’s intuitive, no-code GUI
Easily integrating the approval process with several data
sources, like Credit Bureau, Revenue Service, National Bureau
of Enforcement, and National Bank of Georgia
Client Story

AREA42’s objectives
Overarching goal: improve international trade
by creating new and innovative technological
AREA42 / [Link] solutions and a fluid, risk-savvy B2B trade
ecosystem
Innovation vehicle focused on B2B trade
To achieve that, AREA42 has been looking for
& financing a decision engine with the following features:
Part of one of the largest credit insurance
Ability to make quick, completely automated
groups in Europe decisions
Website: Well-designed GUI for business users
[Link] Modern technology stack and features such
as easy data integrations, model deployment,
[Link]
out-of-the-box simulation layer, etc.

Why has AREA42 chosen

TaranDM?
Simple and flexible deployment of scoring
models, including the support of multiple
analytical environments for the model
development
“In our pursuit of dismantling impediments Both real-time and batch decisioning-making
in B2B trade, we are diligently crafting and processing capabilities
Easy maintenance of the decision engine
an extensive array of financial products
Modern technology stack & “Python-native”
necessitating diverse strategies. Together with
nature of TaranDM
our expansive geographical reach, there Taran team’s credit risk and data science
is a substantial demand for adaptability domain expertise which has ensured smooth
in our decision-making processes and the tools & well-executed TaranDM implementation
we employ to ensure scalability and facilitate
the burgeoning growth of our enterprise. Implementation highlights
Through our collaboration with Taran, we have
acquired a cutting-edge decision engine. Development of a credit bureau adapter
This advanced technology positions us at Deployment of the TaranDM Antifraud module:
In-built data source which stores &
the forefront of our industry. It empowers us
automatically calculates concentration checks,
to exercise precise control over automated cross checks and other aggregations which
underwriting strategies, facilitates swift and are crucial for fraud prevention
uncomplicated model deployments, seamless Examples: number of loan applications from
integration of data sources, and furnishes our the same IP address over past 1h/24h/7d/30d;
clientele with real-time decisions within milliseconds.” number of loan applications from the same
device over past 1h/24h/7d/30d; etc.
Matej Urban, Head of Risk & Data Includes hundreds of pre-defined aggregations
and it’s customizable further (client can create
additional aggregations)
Client Story

Business challenge
& objectives
Product catalog in size of billions of items
iPrice Group (hundreds of gigabytes of raw data), 8 mil.+
Founded: 2014 sellers, 1000+ merchants
The largest e-commerce price aggregator How to rank the products & maximize the
conversion rates?
in Southeast Asia
How to run the product scoring process daily
Active in Singapore, Malaysia, Indonesia,
& efficiently, given the sheer size and complexity
Philippines, Thailand, Vietnam and Hong Kong
of the data?
HQ: Kuala Lumpur, Malaysia
Website: [Link]
Taran has exceeded
all expectations
A robust and scalable big data platform which
requires minimal maintenance, and which has
been used in production since 2019
A state-of-the-art ML model for product scoring,
which has led to 20% increase in revenue/
conversions

The solution is leveraging

the latest and best-perfor-
ming big data technologies
“They are super smart
– in a nutshell. You have The pipelines include daily dump of product
catalog, click logs and transaction logs into
the confidence that they Amazon S3, processing the data using Spark
on Amazon EMR, calculating aggregates and
really solve the problem executing gradient boosting predictive models

in the best possible way” on PySpark

Analytical playground connected with Git
Would you recommend available on demand (in cloud, on click), servers
provisioned daily using Airflow and Terraform
Taran to anyone else? Daily, Taran solution processes 600 GB+ data

“120% yes” in less than an hour and calculates scores for

billions of catalog items

Heinrich Wendel, Co-founder & CPO/CTO

To find out how Taran can assist you in your business,

contact us at info@[Link]
TaranDM
Optional Modules
TaranDM Antifraud

Mitigation of the fraud risk in real-time online processes

Stores & automatically calculates, in the real time, various concentration checks, cross checks and other
aggregations which are useful for preventing frauds
Examples:

number of loan applications from the same IP address over past 1h/24h/7d/30d
number of loan applications associated with the same phone number over past 1h/24h/7d/30d
number of loan applications from the same device over past 1h/24h/7d/30d
how many different addresses has a customer used in the past
…

Fully customizable, client can create additional aggregations

TaranDM Decision Optimizer

Real-time optimization module solving complex business tasks such as revenue

or profit maximization with respect to business constraints / KPIs

Business problem:
TaranDM
How to tackle various product- and Default rate model
client-driven constraints & which depends on interest rate, higher interest can lead to higher probability of default
Optimal decision
product parameters to offer? Propensity model across all
possible interest rates between 5.25% to 8% p.a. models probability of accepting the offer, higher credit amounts and lower Decision the business
maturity of 12 months, 48 months or something in between?
interest rates increase this probability optimization considerations
credit amount in the USD 5,000 to 20,000 range?
Churn/early repayment model
higher interest rates increases the likelihood of client refinancing elsewhere,
thus lower profit on our side

Traditional approach: TaranDM approach:

The optimization algorithm is run on historical data. No need to build the extrapolation models. Portfolio constraints are converted to objective function penalizations using advanced
An extrapolation model for real-time use is built on top of those decomposition algorithms. The optimization algorithm runs in the real-time
results. Thus, no true optimization happening in the real-time Supports full set of optimization methods (linear, mixed-integer, non-linear, constrained, …) via both commercial and free solvers
The optimization software is usually a standalone product, Decision Optimizer is embedded part of TaranDM Decision Flow. Users can run rules or decision tables after the optimization process,
performing the optimization task only perform manual overrides to selected segments, etc.
TaranDM Analytics

Data science platform for automated development of predictive models

Can be used for development of predictive models of all kinds:

application scorecards
behavioral scorecards
ECL scorecards
CRM propensity models
collections segmentation models
…

Full variety of AI/ML modelling methods supported (logistic regression, XGBoost, random forest, …)
Deployment of models to TaranDM via few clicks with model documentation automatically generated
Jupyter-based, further customizable by the client
Our Expertise
in Credit Risk
Management
 Data strategy
Modern Data Sources Provides Additional Predictive Power

Digital Data External Data Internal Data

Device Data Behavioral Data Mobile App Data Telco data Transactional
device type - device finger- copy pasting - typos, accuracy installed apps - timezone data
print - webgl report rate - biometric predictors - geolocation

PSD2 Social circle

Weighting/Blending Optimization

OUTCOME:

10-60% risk cost reduction

 Scorecard development
ML Techniques Can Boost Performance Significantly

Vintage 60+
Logistic regression performing well Default rate

4.5%
on small datasets or when expert
4.0%
knowledge needed
3.5%

3.0%
However, on bigger portfolios,
2.5%
techniques such as XGBoost have 2.0%
significantly better performance 1.5%

1.0%

1.5%

1.0%
3 4 5 6 7 8 9 10 11 12 13
MOB

LogReg default rate XGBoost default rate

OUTCOME:

15-20% risk cost reduction, stable approval rate

 Antifraud
Modern Approach 2-3× More Powerful than Traditional
Simple hard checks:
too many - hard to manage - low hit rate

Better approach:

Digital Data based Cross-, Concentration-

and Other Checks
# applications from given IP/device/... in past 1h/1d/7d/30d
IP anonymizers (Tor, VPN, ...), IP abroad, etc.
Device Behavioral Mobile App External
Data Data Data Data
Automated Antifraud Score

OUTCOME:
up to 3× more accurate fraud detection, reduction of manual activities by up to 80%
 Approval process review & implementation
Improving Business Metrics via Comprehensive Assessment

Data strategy & data sources review

Data sources cost optimization
Approval process streamlining & automation
Lifetime profitability assessment

OUTCOME:
increase in approval rate by 5-20%, increase in profitability by 20-40%
 Data architecture design & implementation
Customized, State-of-the-Art Design 100% Fitting Your Needs

Our core principles:

Focus on real-time performance
Open source components, no vendor lock-in
Raw data & single predictor catalog for both analytics and production
Easy model deployment
Microservices decomposition

OUTCOME:
significant profit improvement opportunities from the ability to analyze
data instantly & react immediately
Contact us
Do you have any questions?
We will be happy to answer them

info@[Link]

[Link]
[Link]