Smriti — One place to ask your whole business

Private beta

Bring your data together and ask anything in plain words. Answers in seconds, from your own numbers, with the working shown.

ChatLive
ReportsSoon
DashboardsSoon
AlertsSoon
Areas of concernSoon

Smriti

Which region missed target last month?

SQL

SELECT region, target_gap_pct
FROM sales_monthly
ORDER BY target_gap_pct ASC;

South missed target by 8.4%. Distributor returns and slower repeat orders drove the gap.

Region	Gap to target
South	-8.4%
West	-2.1%
North	+1.6%

Speaks your business

targetthe monthly planmissedactuals below plan

Answered from SQL

Smriti

Who hasn't paid in 60 days?

3 accounts are past 60 days, with ₹18.4L outstanding.

Account	Days overdue
Anand Traders	72
Mehta & Co.	65
Surya Retail	61

From your ledger

Time to insight

From days to seconds.

The same question, the old way and with Smriti. One of these takes an afternoon. The other takes a sentence.

The old wayHours to days

Export from each tool
Open a spreadsheet
Clean and pivot
Build a dashboard
Wait on a ticket

With Smriti

Seconds

Ask in plain words
Read the answer

Watch it work

One question, from a sentence to an answer you can audit.

The kind of question your dashboard can't answer for you, followed from plain words to the exact query and the proof behind it. No made-up numbers.

Ask in plain words.

No dashboard to find, no SQL to write. Ask the question you would have sent to an analyst.

It speaks your business.

Smriti maps your words to the metrics and tables your team already defined.

Backed by a real query.

The number comes from SQL run on your data, never invented by the model.

The answer, in seconds.

A clear result you can read, compare, and follow up on in the same thread.

See the proof.

The chart and the query behind every answer, so you can trust the number.

Which dealers make up 70% of our primary sales?

Resolved from your business knowledge

primary salesqty dispatched, invoiced onlymake up 70%running cumulative share

SQL

SELECT d.name, sum(x.qty) AS sales,
  sum(sum(x.qty)) OVER (ORDER BY sum(x.qty) DESC)
    / sum(sum(x.qty)) OVER () AS cume_share
FROM dispatches x JOIN dealer d USING (dealer_id)
GROUP BY d.name
QUALIFY cume_share <= 0.70;

8 dealers make up 70% of primary sales. The top three already reach 56%.

Dealer	Cumulative
Verma Distributors	22%
+ Northern Agencies	40%
+ Ganpati Traders	56%

Cumulative share computed in-query · grounded in your business knowledge · scoped to your workspace

Not a dashboard

You just watched an operator work.

A dashboard waits for you to look. An operator takes the question, finds the right tables, writes the query, shows its working, and hands back the answer. That is what Smriti is: an intelligence operator on your data. You direct it; it does the legwork.

One place, not ten

Stop chasing numbers across tools.

Sales in one tool, finance in another, the rest in spreadsheets and an inbox. Smriti brings them together, so one question reaches all of it. No new dashboard, no export, no ticket.

One place to askAsk anything, in plain words

Getting started

Start with the data you already have.

No warehouse, no setup project. Bring a file, and Smriti is ready to answer.

01
Bring your data
Drop in a CSV, Excel, or Parquet file. No pipeline to build.
02
Smriti learns it
Your tables and the words your team uses become a semantic layer you control.
03
Ask anything
Questions in plain words, answered with the SQL behind every number.

Direct connectors to your databases, ERPs, and tools are coming. Today, you start with a file.

Your data lives in Smriti, secured. It stays yours, exportable anytime.

FAQ

Questions, answered.

Do the numbers come from the AI model?

No. Smriti writes SQL, runs it on your data, and returns the result. You can read the query behind every answer.

Do we need a data warehouse first?

No. You bring data into Smriti and it does the analytical storage and querying itself, so there is no separate warehouse to set up. Start with CSV, Excel, or Parquet; direct connectors to your databases and tools are coming.

Can the data team control definitions?

Yes. Your team owns the tables, joins, and metrics Smriti uses, and every answer stays traceable to them.

Who should use Smriti first?

A team that asks repeat data questions every week: sales, finance, operations, or the data team that runs their request queue.

स्मृति

smriti Sanskrit · memory

It remembers how your business talks.

Put an operator on your data.

One operator, growing across your whole intelligence workflow. The question you ask every week becomes the report, the dashboard, the alert.

ChatAsk anything, get the SQL-backed answerLive
ReportsScheduled and deliveredComing
DashboardsLive views of what mattersComing
AlertsTold the moment it changesComing
Areas of concernIt watches your data and flags what to look atComing

Oneplacetoaskyourwholebusiness.