Data Tranquil Inc.
Certified California Small Business. Data, analytics and AI delivered as fixed-scope work.
We build data platforms, analytics and AI systems for organisations that need them to work in production — not as a pilot, and not as a slide deck. Certified by the Department of General Services as a Small Business and a Microbusiness, serving all 58 counties.
Capability
What we can take off your desk
Six problems state departments keep having. Each is a discrete piece of work that can be scoped, priced and finished.
Mandated reporting
Stop rebuilding the same report by hand
Where staff rebuild the same mandated or public-facing report each cycle from exports and spreadsheets, we build the warehouse, pipeline and dashboard underneath it — so the figure is current, repeatable and traceable to its source when someone asks where it came from.
Modernisation
Move off a legacy system without losing the record
Data migration off ageing departmental systems, including the reconciliation that proves what arrived matches what left. In government the history is often the statutory obligation, so this is the work that decides whether a migration is finished or merely cut over.
Interoperability
Connect systems that cannot talk
Application and data integration across platforms that were never designed to meet — between divisions, between a department and a vendor system, or across agencies where a shared view is needed but a shared system is not coming.
Audit readiness
Know why two reports disagree
Data quality, lineage, validation and reconciliation. When the same measure produces different answers in different places, this establishes which is right, documents why, and stops it recurring.
Case and intake
Take a manual document process off caseworkers’ desks
Intelligent document processing, classification, extraction and retrieval applied to applications, case files and correspondence — with a human checkpoint wherever the decision affects a member of the public.
Assessment
A defensible answer before you spend
Feasibility studies, readiness assessments and sequencing, delivered as a written decision document that stands up in a procurement file. Including, where it is the right answer, a recommendation not to build.
We also carry ongoing support and maintenance — for what we build, or for what someone else built and left.
Service lines
Named services, mapped to the codes you requisition under
Data Tranquil Inc. is registered on Cal eProcure under 94 UNSPSC classifications. These are the service lines behind them, so a requisition can be matched to a capability without a phone call.
| Service line | What it covers | UNSPSC |
|---|---|---|
| Data engineering and integration | Pipelines, ingestion and transformation, ETL and ELT, change data capture, application and data integration, data conversion and standardisation. | 81112000 · 81112001 · 81112002 · 81112007 · 81112010 · 43232403 |
| Data warehousing and business intelligence | Warehouse and lakehouse builds, dimensional modelling, semantic layers, reporting, dashboards and the metrics definitions underneath them. | 43232314 · 81112003 · 81111806 · 43232300 |
| Solution and enterprise architecture | Architecting the solution end to end: target-state design, platform and integration architecture, migration sequencing, and the build-versus-buy decision written down with its reasoning. | 81111503 · 81111500 · 80101504 |
| Application development | Building the application itself — custom web applications, internal tools, portals and the services behind them, delivered into your environment rather than handed over as a specification. | 81111500 · 81111504 · 81111507 · 81111508 |
| Agentic AI systems | Autonomous and semi-autonomous agents that carry out real work against your systems: scoped tool access, validated actions, evaluation and guardrails, with a human checkpoint where the decision warrants one. | 43231500 · 43231511 · 81111500 · 81111504 |
| ERP and CRM data | ERP and CRM data integration and migration, application implementation services, and ongoing software maintenance across enterprise business systems. | 81111507 · 81111508 · 81112211 · 81112212 |
| Data governance, quality and access | Data quality, lineage, validation and reconciliation; data classification; access and identity management; audit support. | 81112009 · 81161500 · 81161501 · 81161502 · 43231516 |
| Applied AI and automation | Intelligent document processing, classification and extraction, information retrieval, conversational interfaces, workflow and business process automation. | 43231500 · 43231511 · 83121603 · 83121605 |
| Cloud platform and migration | Cloud data platform build and migration, software and platform as a service delivery, cloud-based data access and sharing, data archiving. | 81162000 · 81162100 · 81162103 · 43233513 · 80161506 |
| IT consulting, assessment and strategy | Feasibility studies, readiness assessments, strategic planning and sequencing, risk management consulting, project monitoring and evaluation. | 80101500 · 80101504 · 80101507 · 80101508 · 80101510 · 80101601 · 80101606 |
| Support and maintenance | Ongoing operation, enhancement and support of what we build, or of an existing system another vendor left behind. | 81112200 · 81112205 · 81112208 |
These are engagements to design and build a working system, not to supply a role by the hour. You are buying an architected solution, an application, an agent or a pipeline — scoped, priced and delivered as a piece of work with an end date. Data Tranquil performs and supervises that work directly.
The full registration covers 94 classifications across software, data services, IT consulting, cloud delivery and access management. If your requisition sits under a code not listed here, ask — it is likely already on the profile.
Agentic AI
Agents that do the work, not chatbots that describe it
California is early in buying this. That is the reason to have the conversation now rather than after the market sets its expectations.
An agent is only useful in government if it can be trusted with a real system and held to account afterwards. What we build is scoped to specific tools, validated on every action, logged so each step is attributable, and stopped at a human checkpoint wherever the decision affects a member of the public or a statutory obligation.
Intake and triage
Read the incoming file and route it
Applications, correspondence and case files classified, key fields extracted and checked against the record, then routed — with anything uncertain escalated to a person rather than guessed at.
Retrieval
Answer from records spread across systems
Retrieval across repositories that share no index, returning an answer with its sources attached so staff can verify it rather than take it on trust.
Monitoring
Watch the data and raise the exception
Agents that run reconciliation and quality checks continuously and raise what has broken, instead of a report nobody reads until an audit.
Drafting
Prepare the first version for a person to sign
Responses, summaries and recurring reports drafted from the source record and handed to staff to review, correct and own. The signature stays human.
Legacy reach
Operate a system with no modern interface
Agents that work against systems which never received an API, so a replacement programme is not the precondition for automating a process.
Internal tooling
Build the small tool nobody has time to build
The internal application a team has been asking for and that has never cleared the project queue, built quickly because the mechanical work is carried by the harness.
None of this requires a department to adopt a platform or sign a multi-year commitment. Each of the above is a bounded piece of work that can be scoped, priced and delivered — and, where the honest answer is that an agent is the wrong tool, we will say so and build the simpler thing instead.
Procurement
How to contract with us
Most vendor sites leave this for the buyer to work out. Here it is plainly.
| Path | Value | What it means |
|---|---|---|
| SB/DVBE Option | $5,000 – $250,000 | Direct award to a certified small business, exempt from advertising. For most departments the fastest route from requirement to signed contract. |
| Fair and Reasonable | Up to $10,000 | For small, well-defined pieces of work. |
| Bid preference | Five percent | Applied on applicable competitive solicitations. |
| Subcontracting | Any value | Our certification counts toward a prime contractor’s small business participation goal. |
| Reciprocity | Varies | Our DGS certification is recognised by California government entities operating reciprocity arrangements. |
Every engagement is fixed-scope and fixed-fee. The fee is set on a short scoping call once the requirement is confirmed, because an honest scope depends on the systems involved and the state of the data. Tell us the requirement and you will have a written scope and a price within two business days.
Where to start
Start with a discovery, not a build
Departments often know the problem before they can write a scope. Quoting a fixed price against an undefined scope is how projects go wrong for both sides.
So we start with a fixed-scope, fixed-fee discovery. We look at your data, your workflow and one candidate use case, and give you a written recommendation covering what is worth building, what is not, and what has to be fixed first.
- 01
You tell us the requirement
One use case, one workflow, or one system that is not doing what it needs to do.
- 02
We look at the data and the workflow
What exists, what condition it is in, and what would actually have to change.
- 03
You get a written recommendation
Scope, sequence, risks, and a price for the build if there should be one. If the answer is that nothing should be built, that is a real and useful result.
Compliance
AI governance and Executive Order N-5-26
Executive Order N-5-26 requires AI vendors doing business with California to attest to and explain their policies and safeguards. We answer with mechanism, not policy language.
How do you reduce the risk of harmful bias?
An evaluation layer with source traces and freshness checks sits around every model call. Systems fail closed on missing context rather than guessing. Agent actions are deterministic, validated calls rather than free-form generation, and human checkpoints sit wherever the decision warrants one.
How do you protect civil rights and civil liberties?
Least-privilege data access is enforced per call, organisation scope is isolated, and provenance records make every action attributable and reviewable after the fact. The systems we build carry no surveillance functionality.
How do you handle the risk of illegal content?
The systems we build are enterprise workflow platforms rather than generative media producers, and carry no user-generated media distribution surface. Model providers are commercial and vetted, with their own content safeguards.
Our written AI Governance and Safeguards Statement is available on request.
Key personnel
Public sector and regulated-industry experience
Data Tranquil Inc. was incorporated in 2023 and its own delivery record is short. We would rather say so than imply otherwise.
What sits behind the work is more than 25 years of the founder’s hands-on delivery, earned as a consultant employed by systems integrators and as an employee — not as Data Tranquil engagements. Government evaluates this separately, as key personnel experience, and that is exactly what it is.
| Organisation | Sector | Delivered through | Work delivered |
|---|---|---|---|
| SIDBI — Small Industries Development Bank of India | Government development bank | HCL | End-to-end business intelligence on a custom banking application: dimensional modelling, OLAP reporting, application server and directory configuration. |
| IAEA — International Atomic Energy Agency | United Nations agency | Zensar | End-to-end business intelligence and data warehouse delivery — OBIEE, OBIA, DAC and Informatica. |
| Phoenix Park Gas Processors | State-linked gas processor, Trinidad & Tobago | HCL | Reporting implementation across financial, project, asset management, procurement and HR domains. |
| Rosendin Electric | Electrical infrastructure contractor | Zensar | Business intelligence platform upgrade, OBIEE 10 to OBIEE 11. |
| Sunrun | Regulated residential energy | Direct employee | Enterprise data and marketing analytics at national scale, across a multi-channel sales and installation operation. |
Each engagement was delivered as a consultant on the named firm’s team, or as a direct employee — the “delivered through” column says which. They are listed as key personnel experience and are not Data Tranquil Inc. customer relationships or endorsements.
Delivery
How a small firm delivers at scale
A fair question, and one worth answering with evidence rather than assurance.
The mechanical work is carried out by an engineering harness — evaluation, guardrails, tool contracts and observability wrapped around the model. The judgment is done by an engineer with thirty years of delivering enterprise data and AI programmes hands-on. Data Tranquil performs and supervises all work directly and does not subcontract delivery.
That is a claim worth evidence, so here is some. We hold our own build work to a documented verification discipline, and the records are public. Every claim on this site has to pass an automated check before it can ship — the build fails on anything we cannot substantiate — and an independent review reads the work before code is cut. That review caught a stale-base error on this build before it shipped. The record is in the commit history.
Data Tranquil also builds and operates its own production AI system under the VexioHQ brand — a voice agent that handles live calls and books appointments against real calendars. It matters here for one reason: we carry the operational load of a production AI service ourselves, including uptime, integration with third-party systems, and what happens when a model or a dependency fails. That is the same class of problem a department inherits the day an AI system goes live.
We hold ourselves to the standard we would be held to.
| Phase | Artifact the department receives | Acceptance |
|---|---|---|
| Discovery | A decision memo: baseline, risks and feasibility, and a scoped recommendation — including a recommendation not to build where that is the honest answer. | The department confirms scope before any build begins. |
| Pilot | A working workflow built against departmental data, with an evaluation record measured against criteria agreed before the build. | The department accepts the result or stops; the next phase requires separate authorization. |
| Embed | Production release, handover of operating knowledge, a named departmental owner and support arrangement, and a value review against the baseline. | Signed acceptance by the department’s named owner. |
The same delivery approach applies to every engagement, public sector or private.
Verification
Certification and registration
Every detail below is verifiable in the Cal eProcure certified firm database.
| Field | Value |
|---|---|
| Legal name | Data Tranquil Inc. |
| DGS certification | Small Business (SB) · Microbusiness (MB) |
| Certification ID | 2055176 |
| Cal eProcure Bidder ID | BID0134304 |
| Service area | Statewide — all 58 counties |
| Business type | Service |
| Principal office | California — address of record on file with the CA Secretary of State, entity no. 5735127 |
| CA Secretary of State entity no. | 5735127 |
California
Tell us the requirement.
A real person reads every message. You will have a written response within two business days.