Data and AI stack: delivered work and growth capabilities
Netbase's data and AI stack covers recommendations, knowledge retrieval, document AI, model operations and workflow automation. 4over4's recommendation engine is delivered; so are RAG, document AI and MLOps projects for unnamed clients. Automation modules are productized; other AI areas are growth capabilities.
Review technology fit See related workReviewed by David (CEO) · Updated 15 Sep 2026
As reported by 4over4, revenue grew 82% within six months
Design-file production time fell 40% at 4over4
Fulfilment time fell 50% at 4over4
What this family covers, and why buyers consider it
Data and AI tooling turns order history, images and messy inputs into actions software can take. This page covers stack fit and trade-offs; the engagements are agentic AI automation and AI integration in the AI & Data family.
AI maturity labels
Productized: a Netbase module client projects can start from.
-
In delivered work
Recommendation engine
Shipped for 4over4.
-
Available capability
Machine learning, NLP, computer vision, generative AI and AI with IoT
Offered, but not yet tied to a published Work item.
The stack, by job and maturity
-
Image and vector conversion
Growth capability- Component
- Raster-to-vector conversion of customer artwork
-
Recommendations
Delivered (4over4)- Component
- Recommendation engine based on browsing and purchase history
-
Knowledge retrieval
Delivered (anonymised client)- Component
- RAG knowledge assistant
-
Document reading
Delivered (anonymised client)- Component
- Document AI platform
-
Model release and monitoring
Delivered (anonymised client)- Component
- MLOps pipeline
-
Messaging assistants
Delivered (client not named)- Component
- WhatsApp AI chatbot with CRM sync
-
Content moderation
Delivered (client not named)- Component
- AI filtering of offensive ads
-
Workflow automation
Productized- Component
- Workflow automation toolkit
-
Conversational assistance
Productized- Component
- AI chatbot and WorkChat integrator
-
Prediction and text understanding
Growth capability- Component
- Machine learning and NLP
-
Image and document understanding
Growth capability- Component
- Computer vision
-
Drafting and content
Growth capability- Component
- Generative AI
-
Connected devices
Growth capability- Component
- AI with IoT
When to choose it, and when not to
-
Rules-based automation
- Choose it when
- Inputs are structured and the rules are known
- Avoid it when
- Inputs arrive as free text, PDFs or images that rules cannot read
-
AI file conversion
- Choose it when
- Customers upload artwork that staff fix by hand at volume
- Avoid it when
- Volumes are small or every file needs a designer's judgement anyway
-
Recommendations
- Choose it when
- Enough browsing and order history to learn from
- Avoid it when
- A small catalogue or thin traffic
-
Generative AI and agents
- Choose it when
- Reading, drafting or triage steps vary and a person can review the output
- Avoid it when
- Decisions on money, contracts or people would run unreviewed
Implementation patterns we use
A model is added only for inputs rules cannot handle.
Low-confidence results and customer-facing actions go to an approval queue.
Time per task, error rate and conversion are recorded before the build and compared after it.
Each integration defines which fields move and why.
Trade-offs to weigh
-
Cost
Model and inference charges grow with volume; rules cost little to run
-
Accuracy
Models are probabilistic, so confidence thresholds and review queues are part of the design
-
Data
Recommendations need clean product and order history
-
Security
New risks such as prompt injection, sensitive information disclosure and excessive agency (OWASP Top 10 for LLM applications, 2025)
-
Talent
Data and machine-learning skills are scarcer; productized modules reduce what must be built
-
Maintenance
Models drift and model interfaces change; accuracy is re-checked after launch
Security and operations
AI features follow Netbase delivery practices: secure code review and version control, TLS in transit and AES at rest, role-based access control, MFA for admin dashboards, vulnerability scanning and penetration testing, and disaster recovery. For governance we map risks against the four functions of the NIST AI Risk Management Framework, and each automation keeps a manual fallback.
How Netbase applies it across services and solutions
The commerce operations automation solution is mainly rule-based order, file and routing automation, with AI at specific steps. Printcart, a Netbase Business Division, shows that foundation at product scale: it turns online orders into print-ready files and routed fulfillment. To begin, read our guide to AI automation for business operations.
Versions, ecosystem and verification
-
SVG output format
File automation for 4over4, not AI- Vendor
- W3C (open standard, SVG 2)
-
Recommendation engine
Delivered- Vendor
- Built by Netbase for 4over4
-
Python for data processing
In the stack; Python developers hired- Vendor
- Python Software Foundation
-
Workflow automation toolkit; AI chatbot and WorkChat integrator
Productized- Vendor
- Netbase productized modules
-
Language models
Works with, chosen per project- Vendor
- OpenAI, Anthropic (Claude), Google (Gemini), Meta (Llama) and other open-weight models
Netbase claims no AI or data vendor partnership, tier or certification. Verified on 2026-09-15; projects pin their versions.
Work example, expert and next steps
- 4over4 e-commerce conversion. Netbase built a recommendation engine and automated the conversion of Adobe Illustrator (.ai) files to SVG. As reported by 4over4, revenue grew 82% within six months, design-file production time fell 40% and fulfilment time fell 50%. The results reflect the whole project, not the recommendation engine alone. Read the 4over4 case.
Expert owner: David, Chairman, founder and CEO of Netbase, reviews this page and AI scope decisions on client builds.
Compare families on the technologies overview, or send one workflow and a data sample, and we will review technology fit.
4over4 reported 82% higher e-commerce revenue and a 48% higher conversion rate.
Keep ReadingPlan the next step with a Netbase consultant
Where Netbase applies this stack
AI automation and agents that keep people in charge
Netbase provides AI automation and agent development for operations teams that want repetitive, multi-step work done by software while people keep approval over the decisions that matter. We combine rule-based workflow automation with AI steps where they add value, design the human approval points in, and measure return against a baseline taken before the build.
Learn More
AI-assisted ecommerce operations automation: from paid order to delivered parcel
Ecommerce operations automation removes the manual order, production and fulfilment steps behind a storefront: routing, file preparation, status updates and back-office sync, with AI flagging risky orders and triaging exceptions. It is for merchants and online printers whose order volume has outgrown their staff. In delivered work, 4over4 cut design-file production time 40% and Geo-Tek cut average order processing time 30%.
Learn More
Related insights
Start AI automation with one high-volume, checkable process, not a platform. Measure the baseline first, automate the predictable steps with rules, add AI only where inputs are unstructured, keep people approving money, customer-facing and irreversible actions, and judge success by time, errors and cost per case against that baseline, with security designed in.
Discuss a project
Netbase JSC helps organizations design, build, modernize, and operate digital products and AI-enabled business systems.+84 937 869 689
91 Nguyen Chi Thanh, Dong Da, Hanoi, Vietnam
Get in touch
Tell us what you want to build, modernize, or operate.