Platform · Deployment and governance

Run agents where your data must stay, with clear limits and a record of every action.

Name: AksharAI
Brand: AksharAI
Price: 7500 INR

Choose where AksharAI runs: hosted, in your private cloud, on your own servers or under your own brand. Then set who can use each agent, what data it may touch, how it speaks and which actions it may take, with a log your auditors can read.

Book a demo See the agents

AksharAI / board

AksharAIRelease board: HR copilotSearchAK

Release board: HR copilot

Drafted2

Payslip questions
Shift swap requests

IT review2

SSO groups for plant 3
Log export to SIEM

Risk sign off2

Salary data rule
Leave apply action

Live2

Leave balance answers
Policy questions

Representative screen with sample data. Client screens stay private.

The problem

The pilot worked. Then risk and IT asked questions

A manufacturing group in Vadodara built an HR copilot that answers leave, payroll and policy questions for 4 plants. The pilot was popular. Before wider release, the CISO asked where employee data is stored, the HR head asked who can see salary answers, and internal audit asked how they would check what the copilot told people. Without clear answers the project stalled for two months.

What it gives you

Inside deployment and governance

Hosted SaaS, private cloud, on premise or white label

Start hosted for speed, run in your own cloud account or data centre when data must stay with you, or offer the platform to your clients under your brand.

Access and visibility

Decide which teams, roles or individuals can use each agent and which agents they can even see, using your existing single sign on where possible.

Data permissions

Limit each agent to the records the user asking is allowed to see. A plant supervisor's question about leave never returns another plant's salary data.

Prompt and tone control

Lock the instructions that set an agent's behaviour so only named people can change them, and keep the tone consistent with how your organisation speaks.

Action boundaries

List what an agent may do, cap amounts and counts, and require a person's approval for anything that changes money, contracts or employee records.

Logging and oversight

Every question, answer, source and action is logged with user and time. Reviewers can search, sample and export the log for audit or a grievance.

How it works

The order you work in

Step 1Choose where it runsAgree the deployment option with IT based on the data involved, your cloud setup and any client or regulatory requirements.
Step 2Write the rulesWith the business owner, set who may use each agent, what it may see and do, and what always needs a person.
Step 3Review before releaseIT, risk and the business owner each sign off on the agent's settings and test results in the platform.
Step 4Watch it in useSample conversations each week, review blocked actions and alerts, and adjust the rules as you learn.
Step 5Review access regularlyRecheck users, connections and permissions every quarter, and when people change roles or leave.

AksharAI / list

AksharAIActivity log, 9 Oct 2026SearchAK

Activity log, 9 Oct 2026

Time	User	Request	Rule	Outcome
10:14	Suresh K., Plant 2	Leave balance	Own records	Answered
10:21	Pooja M., Plant 1	Apply 2 days leave	Action allowed	Confirmed
10:37	Imran S., Plant 3	Colleague's salary	Data permission	Blocked
11:02	Deepa R., Plant 4	Maternity policy	Policy source	Answered
11:18	Harsh V., Plant 2	Change bank details	Needs HR approval	Escalated to HR

Representative screen with sample data. Client screens stay private.

An example

An HR copilot on private cloud

The group deploys AksharAI in its own cloud account in an Indian region, so employee data stays in its environment. The copilot answers from the HR policy and each employee's own leave and payslip records, using single sign on to know who is asking. It can apply for leave on the employee's behalf but cannot approve it, and it never discusses anyone else's pay. Internal audit gets read access to the log.

An illustrative example with sample data.

Questions

About deployment and governance

Which deployment option should we choose?

Hosted is quickest for a pilot with low risk data. If agents will handle personal, financial or health data, most organisations choose private cloud or on premise so data stays in their environment. We help you decide based on the data each agent will handle.

Does using AksharAI make us DPDP compliant?

No tool does that on its own. AksharAI gives you the controls: data permissions, logs, deployment in your environment. Notices, consent and data requests sit with your organisation; TrustOS, the group's DPDP product, can help there.

Can we offer agents to our own clients under our brand?

Yes. The white label option lets you run the platform under your name and domain, with separate workspaces and data for each of your clients.

Keep going

Bring one slow, repetitive process. We will show you an agent doing it.

A working session with someone who builds these every week.