Emails, forms, contracts and reviews pile up every day.
Software picks out names, dates, amounts and the topic.
Urgent items go first and routine ones are filed for you.
Anything uncertain is set aside for a person to review.
NLP Development
Software that reads and sorts your text for you.
Natural language processing (NLP) lets software understand written text. We build systems that classify, extract and analyse information from emails, forms, contracts, reviews and reports.
What you get
01Less manual readingIncoming text sorted, tagged and routed without staff opening every item.
02Structured dataNames, dates, amounts and other details pulled out of documents into your systems.
03Clearer insightTrends and issues found across large volumes of feedback, tickets and reports.
How we deliver
Four stages, each ending with something you can see and review before the next one starts.
01
Sample
We collect real examples of your text and agree what needs to be found or sorted.
02
Annotate
Your experts and ours label examples, which become the standard the system must meet.
03
Develop
We build and test models, choosing the simplest approach that meets the standard.
04
Integrate
We connect the system to your inbox, document stores or software and track accuracy.
What we buildOur NLP development capabilities
Text classification
Sorting emails, tickets and documents by topic, urgency or department automatically.
Discuss thisInformation extraction
Pulling names, dates, amounts and clauses from contracts, invoices and forms.
Discuss thisSentiment and feedback analysis
Measuring how customers feel and grouping common complaints and requests.
Discuss thisSearch that understands meaning
Search that finds relevant content even when people use different words.
Discuss thisTranslation and multilingual
Processing and translating text in several languages, including regional ones.
Discuss thisDocument comparison
Spotting changes and missing clauses between contract versions or policy documents.
Discuss thisRedaction of personal data
Finding and masking names, ID numbers and other personal details in documents.
Discuss thisText processing pipelines
Reliable flows that take text in, process it and send results to your systems.
Discuss thisKlasendra
Decades of seismic graphs existed only on paper, so the archive could not be analysed.
Selected work · KlasendraAnimated illustration
Energy sector: document data extraction
What we built
- A specialised OCR system for paper seismic graphs
- Extracts structured data from each graph
- Makes the historical archive available for analysis
What it changed
The energy company can analyse its historical archive for the first time.
Delivered by Hoki and its engineering partners. Product and client names belong to their owners.
Discuss a similar projectFrom the Hoki portfolio
The technology we build with
We pick tools around your existing systems, your team and how the product will be run.
Language models
Hugging Face TransformersBERTOpenAIAnthropic ClaudeLlama
NLP libraries
spaCyNLTKscikit-learnGensimSentence TransformersPython
Search and data
ElasticsearchOpenSearchPostgreSQLApache KafkaAirflowAWS
Where this applies
The same engineering, shaped by the rules and realities of each sector.
Financial services
Extracting terms from contracts and screening documents for compliance issues.
Healthcare
Pulling key details from referral letters and coding clinical notes.
Retail and e-commerce
Analysing product reviews and routing customer emails to the right team.
Logistics and supply chain
Reading invoices, bills of lading and customs forms into your systems.
Energy and utilities
Sorting customer emails and analysing field reports for recurring faults.
Public sector and education
Processing public feedback and applications, and redacting personal data.
Frequently asked questions
Can it handle our languages?
Most widely used languages are well supported, and many regional languages can be handled with the right models and examples. We test with your real text before committing to an approach.
How does it fit with our current systems?
We connect to common email platforms, document stores, CRMs and databases. Results can be sent wherever your staff already work, so they do not need to learn a new tool.
How do you handle personal data in documents?
We agree data handling rules first. Processing can run in your own environment, personal details can be masked before analysis, and access is limited to those who need it.
What happens after launch?
We can monitor accuracy, review items the system was unsure about and retrain it when your documents or categories change. The support level is agreed with you before launch.
Another question? Email hello@hoki.co
Let software do the first read
Send us a sample of the text your team handles. We will show what could be sorted or extracted automatically.
Start a project
Tell us what you want to build.
Share a few details about your NLP development project. We will reply with questions and a suggested next step.
- What problem you want solved
- What you already have in place
- Any timing or budget you are working to
Start a project
Tell us what you want to build
A few details help us bring the right people to the first call.
Thank you.
Your enquiry is with our team. Here is what happens next.
1We read your enquiry and bring in the people best placed to help.
2We reply by email to arrange a call at a time that suits you.
3On the call, we agree a clear next step, with scope and cost.