Published

Your application stores orders in Amazon DynamoDB. The operations team wants to find overdue orders above $500, inspect the matching records and see how those orders are distributed by region. It is a straightforward business question, but the application needs a search and reporting path that can answer it as the filters change.
Introducing Exasift, a new product from the team behind FormKiQ, for serverless search and reporting on DynamoDB inside your own AWS account. Exasift is in early access, with a public example and a waitlist for teams interested in the pilot. The focus is finding records across attributes, then counting, grouping and reporting on the records that matched.
An application can begin with well-defined lookups, then acquire new operational questions: which orders are overdue, which exceed an amount, and where are the matches? The combination needed for a report may differ from the one used by the original screen.
DynamoDB’s Query operation uses equality on the partition key, with an optional condition on the sort key. A filter expression is applied after items are read and does not reduce that operation’s read-capacity consumption. A flexible search screen therefore deserves deliberate design around its actual queries and data.
Exasift is being built around that need: a serverless search index and API alongside your DynamoDB tables, with filtering and reporting connected to the same matching set. Its proposed operating model keeps the deployment and upgrade timing under your control, without a separate search cluster to maintain.
Choose the attributes and conditions that identify the records you need.
Check which items satisfy those conditions and which are excluded.
Group and count the matches to answer the operational question.
The interactive example at exasift.com uses ten illustrative orders. Its initial filters are status = "overdue" and total > 500. Six orders match: four in North America and two in Europe. These are sample records and a browser demonstration, rather than production query results or a performance benchmark.
| Region | Matching orders | What the count means |
|---|---|---|
| North America | 4 | Four sample orders meet both the status and amount conditions. |
| Europe | 2 | Two sample orders meet the same conditions. |
| All matching orders | 6 | The regional counts describe one selected set, rather than separate searches. |
Find orders already marked overdue whose total exceeds $500. The example uses the stored status; it does not calculate overdue status from today’s date.
Keep the status and amount filters selected. The page highlights six matching rows among the ten sample orders, so you can inspect the result before reading the report.
An overdue $318.50 order is excluded by the amount condition. A paid $760 order is excluded by the status condition. Both filters matter to the answer.
The report groups those same matches into North America and Europe. You can connect the count back to the individual rows instead of treating the chart as a separate answer.
Remove the status filter while keeping the amount filter. Eight sample orders now match, including paid or open orders above $500. The result and report change together.
Exasift’s published architecture places the DynamoDB tables, serverless search index and search/reporting API inside the customer’s AWS account. Changes from DynamoDB feed the index, and the application queries the API. The intended model avoids a separate export-and-report workflow and lets your team control when its deployment upgrades.
For a pilot, bring the actual application requirements: how quickly updates must appear, which records each user may see, how reports handle a large matching set and what recovery needs to look like. Those requirements need workload validation. The public example illustrates the interaction; it does not establish production limits, latency, cost or immediate visibility of every update.
Our goal is to give FormKiQ customers a lower-cost option for suitable search and reporting workloads, reducing the need for a separate Amazon Redshift or OpenSearch deployment. Think of a document review queue filtered by department, status and renewal date, with counts grouped by document type. The planned Exasift module would bring that kind of metadata search and operational reporting into the FormKiQ deployment.
We are also investigating full-text search in Exasift. Its scope and availability are still being evaluated. The current public example demonstrates attribute filtering and grouped reporting.
The fit depends on the work. Redshift serves broader data warehousing and SQL analytics needs, while OpenSearch supports full-text search and relevance requirements. Where attribute filtering and grouped reports answer the question, Exasift aims to reduce infrastructure and operating costs. Validate the query behaviour, data volumes and total cost against the actual workload during the pilot.
The most useful starting point is a DynamoDB question your team already needs to answer. It might be an operational queue, an order filter or a regional breakdown. A small, well-described workload gives the pilot a clear purpose.
Exasift brings search and reporting together around a matching set of DynamoDB records, with a deployment model centred on the customer’s AWS account. We also plan to make it available as a FormKiQ module for suitable metadata-search and reporting workloads. Explore the example, then bring the question your own application needs to answer to the early-access discussion.
The open-source foundation — API-first, deployable into your own AWS account, and free to use. Right for architecture validation and early implementation.
Production-ready editions for departments and complex workflows. Start with a Proof-of-Value deployment or go straight to production.
For governance-heavy environments with residency, sovereignty, assurance, and multi-jurisdiction requirements. Talk to us about the right deployment model.