Internal chat assistant
Your organisation’s own ChatGPT, on your own servers.
Private LLM Company AI, data stays inside
A language model deployed in your own infrastructure, trained on internal documents and PDPA-compliant. Your team gets ChatGPT-level answers without data leaving the company.
Talk to an engineerWhat we build
In-house AI as capable as the cloud services, with your data kept at home.
Internal document search
Plain-language questions, answers with source citations.
API for your systems
Internal systems call AI capabilities securely.
Features & capabilities
Model of your choice
Llama, Qwen or other open models by budget and language.
Zero data egress
Not a single external API call.
Role-based access
Who sees what, mapped to your org structure.
Audit log per query
Full usage records, reviewable any time.
Fine-tuned on your work
The model gets sharper in your domain.
SSO / Active Directory
Sign in with the corporate accounts you already have.
Teams and industries whose data can’t leave the building
Every team and every industry has its own reason the data cannot leave. These are the systems we install for each.
Private LLM for contract work
“Legal wants AI to read contracts, but every contract holds counterparty data, and pasting it into a public tool breaks policy.”
What we build- A language model installed on your own servers or in your own cloud account, reading contracts uploaded as PDF or Word
- Every clause compared with the company’s standard template, with the differences highlighted and explained
- A plain-language risk summary by topic: payment terms, penalties, termination and governing law
- Search across the whole contract archive in plain questions, such as "which contracts have a non-compete longer than two years", with page references
- An audit log of who read which contract, what they asked and when, for compliance
Connects withMicrosoft 365 · SharePoint · Google Workspace · Active Directory · Document Management · DocuSign
HR assistant on a Private LLM
“Salaries, reviews and employee records are personal data under PDPA, so the team doesn’t dare use AI for HR work at all.”
What we build- An internal chat assistant that answers staff questions from the employee handbook, rules and announcements, citing the clause
- Role-based permissions: staff see their own data, managers see their team, HR sees what the policy allows
- Drafts of announcements, employment letters and benefit answers for HR, built from real data in the HRIS
- Search across records, leave and reviews in plain questions, with every lookup logged
- An audit log the DPO can open to see who asked about whose data and when, as PDPA requires
Connects withSAP SuccessFactors · HRIS · Microsoft 365 · Google Workspace · LINE OA · Active Directory
Private LLM for financial documents
“Pre-announcement financials and budget plans can’t leak a single line, yet the team still summarises hundred-page documents by hand.”
What we build- A model inside the company network that turns financial statements, audit reports and loan agreements into a one-page summary
- Questions across several documents at once, such as "which covenant is closest to a breach", with page and table references
- Drafts of notes to the accounts, memos to management and replies to the auditors from data in the system, for the team to finish
- Budget compared with actuals from the ERP, with the lines that drift from plan flagged and explained from the documents
- Permissions by closing period: pre-announcement data visible only to people on the list, with an audit log on every question
Connects withSAP · Oracle · Microsoft Dynamics 365 · Microsoft 365 · SharePoint · Active Directory
Private LLM for engineering knowledge
“Designs, formulas and source code are the most valuable things the company owns. Send them to someone else’s cloud and who guarantees they won’t be used for training?”
What we build- A model that searches and answers from machine manuals, drawings, repair reports and lab records inside the company, citing the source
- An assistant for the software team that reads your source code, explains what each module does and drafts code without the code leaving the network
- Tuned with the vocabulary, part numbers and units your team uses, so a search finds the part even by its nickname
- Usable on the floor from a tablet on the plant network, with no outside internet connection
- Knowledge captured from senior engineers before they retire, with interviews transcribed and repair notes indexed into the searchable store
Connects withERP · CMMS · PLM · GitLab · File Server · SharePoint · Active Directory
Customer reply assistant on a Private LLM
“Medical history, account details or ID numbers appear in every customer conversation. Using a public AI to help reply means sending customer data out every time.”
What we build- An assistant inside the contact-centre screen that reads the customer’s history from the CRM and past tickets and drafts the reply for the agent to send
- A summary of the customer’s previous conversations for the agent to read before picking up, inside your systems
- ID numbers, account numbers and health data masked automatically, so agents see only what their permissions allow
- Search across policies, product terms and troubleshooting steps in the internal manuals, citing the clause
- An audit log of which agent opened which customer’s history and which drafts were edited before sending, for the DPO
Connects withSalesforce · Zendesk · Freshdesk · LINE OA · Genesys · Active Directory
Private LLM for executives and the board
“Board papers, merger plans and unannounced matters. Executives want AI to summarise them, and no tool on the market is trusted enough.”
What we build- An executive assistant on company infrastructure that turns a board pack of several hundred pages into the points that need a decision
- Connected to executives’ email and calendar, preparing a pre-meeting brief from the attachments and the items left over from last time
- Drafts of board memos, shareholder letters and answers to analyst questions from internal data, for the company secretary to finish
- Data separated between group companies and between document sets: merger papers visible only to the people on the list
- Every access logged, so compliance can see who opened which document before the announcement
Connects withMicrosoft 365 · Google Workspace · Board Portal · SharePoint · Active Directory · Private Cloud
Private LLM for financial institutions
“Staff quietly use ChatGPT to summarise customer documents because it’s fast. Compliance finds out, bans it, and everyone goes back to doing it by hand.”
What we build- A model in the bank’s data centre that summarises, searches and drafts like the public tools
- Permissions by org structure and an audit log compliance can inspect for every question
- Penetration-tested, with documentation ready for internal and external auditors
- An assistant for the credit team that turns a borrower’s financials and documents into one page, with the points that need a follow-up question
- Search across circulars, rules and internal operating manuals in plain questions, for branches and the call centre
Connects withActive Directory · Core Banking · LOS · SharePoint · Kubernetes · Microsoft 365
Private LLM for health data
“Doctors want AI to summarise a patient’s history before the visit, but health data cannot leave the hospital by law.”
What we build- A model inside hospital systems that summarises records and searches internal treatment guidelines
- Role-based permissions: doctors see only patients in their care, nurses and medical records staff see what their role needs
- Designed with the hospital’s legal team to meet PDPA and health-data rules
- Drafts of discharge summaries, referral letters and insurance claim documents from data in the HIS, for the doctor to check and sign
- Search across allergies, past lab results and current medication in plain questions, citing the date and the HIS entry
Connects withHIS · Medical Records System · LIS · PACS · Active Directory · Microsoft 365
Private LLM for clients’ confidential documents
“Our work is entirely clients’ confidential documents. To use AI on contracts, we must be able to tell clients exactly where their data goes.”
What we build- A model on the firm’s servers that searches, compares and drafts from your document library
- Data walls per client and matter: one team cannot search another matter’s files, and conflict-of-interest walls hold
- A usage log you can show clients: who opened their documents and when
- Judgments, old agreements and meeting notes distilled into issues with page references, for lawyers preparing an opinion
- First drafts of agreements and legal opinions from the firm’s templates and past work, for the lawyer to finish
Connects withDocument Management · iManage · NetDocuments · Microsoft 365 · Active Directory
Private LLM for telecom operators
“We have tens of thousands of call-centre conversations a day and network tickets every technician writes differently. We want AI to summarise them, and subscriber data cannot go outside, under our licence and under PDPA.”
What we build- A model in the operator’s data centre that summarises every call-centre conversation into the reason for the call, the outcome and what is still open
- An assistant in the agent’s screen that reads the number’s plan, outstanding balance and usage history by permission and drafts the reply
- Search across network tickets and past outage reports in plain questions, so the NOC finds the recurring fault that was fixed before
- Daily grouping of complaints with the areas where calls are rising unusually, sent to the network and service teams
- ID numbers and personal usage data masked by permission level, with an audit log on every question
Connects withBilling · CRM · Genesys · ServiceNow · Active Directory · Kubernetes
Private LLM for critical infrastructure
“Operating data, maintenance plans and network drawings are critical-infrastructure data that policy keeps inside a closed network, so teams still search manuals in paper binders.”
What we build- Installed in the company’s closed network, working without internet access
- Searches manuals, standards and past incident reports in plain language, citing the source
- Permissions per plant or area, with usage logged to cybersecurity standards
- Drafts of incident reports and maintenance work orders from floor notes and CMMS data, for the shift supervisor to check
- Questions answered from operating data in the historian, such as when this abnormal reading last occurred and how it was resolved
Connects withActive Directory · CMMS · SCADA Historian · File Server · SAP PM · SharePoint
Private LLM for claims and underwriting
“One claim comes with a medical certificate, receipts and ten pages of treatment notes, and a person reads every one. We want AI to help, and it is customers’ health data, so it cannot go outside.”
What we build- A model inside the company’s systems that reads scanned medical certificates, receipts and treatment notes and summarises the claim on one page
- Claim items compared with the policy terms and benefit table, with items outside cover and missing documents flagged
- An underwriting assistant that summarises the medical history from the application and test results, with the points to ask the doctor about
- Claims that resemble past cases found to be irregular flagged and passed to the investigation team for a person to look at first
- Role-based permissions, so health data is visible only to the person handling that case, with an audit log on every access under PDPA
Connects withCore Insurance · Claims System · Microsoft 365 · Active Directory · LINE OA · SharePoint
How it works
01
Infrastructure & model fit
Hardware and model sized to your budget and workload.
02
Deploy & index
System installed, documents connected, security configured.
03
Harden & train
Penetration-tested, team trained, delivered with runbooks.
Works with the tools you already use
Works with the identity systems and document stores you already run.
LlamaQwenvLLMOllamaLangChainElasticsearchActive DirectoryKubernetes
Development process
- 1
Discover
Requirements, users and success metrics, with scope and price fixed before we start.
- 2
Design
UX and system architecture; the prototype is approved before anything is built.
- 3
Build
AI-accelerated sprints with a demo every week, reviewed by senior engineers.
- 4
Test
QA, security and performance verified against the agreed scope.
- 5
Launch & care
Production deploy, team training, and a monthly care plan.
Why DNA Maker
500+ projects since 2012
Enterprise track record across Thai banking, energy and retail leaders.
Fixed scope & price
Agreed before we start, with no surprises mid-project.
AI-fast, senior-reviewed
Our team builds with AI, so we deliver 3× faster with senior engineers guarding quality.
5 languages, Bangkok-based
We communicate and deliver in Thai, English, Japanese, Korean and Chinese, with PDPA in mind.
Turn your business problem into a system that works for you
Tell us today — get an executive-ready proposal with the plan and budget.
Get started