The engineering tax on data,Eliminated.
A governed runtime that turns business intent into production-safe APIs and insights — without writing a single line of backend code. Your schema is the API. Your question is the dashboard.
NO CODE GENERATION
Schema → API in minutes
ACCESS CONTROL BUILT IN
Governed by default
MULTI-DATABASE
Postgres · MySQL · ClickHouse
The Problem
Four delays. None of them necessary.
— Every sprint pays the cost.
EXPOSURE DELAY
Build it safely — wait for backend.
A new table is ready. The business logic is understood. Exposing it safely still takes repeated engineering cycles every time. The capability exists. The API does not. Yet.
ANSWER DELAY
Answers exist — access does not.
Operational answers live in production data. But teams still route through BI tools and engineering queues. Intent in plain language, output weeks later as a dashboard that's already stale.
BACKLOG DELAY
Every request — a new engineering task.
Customer-specific reports pile up. Each request is legitimate. Each becomes a sprint ticket. Engineering builds custom work that cannot be reused, at the expense of product work that should be the priority.
DISCOVERY DELAY
Signals hidden — questions go unasked.
Valuable patterns stay buried because teams can only answer questions they already know to ask. The insight that would have changed a product decision sits untouched in a database with no one watching it.
"These four delays are not isolated inefficiencies. They are symptoms of the same root cause: the gap between business intent and governed, production-safe output has to be crossed by hand, every single time."
What is Minimal?
A governed runtime that turns business intent into production-safe insights and APIs — without the engineering overhead.
No Code Generation
Schema → production API in minutes
Minimal does not write code. It reads your schema and executes intent at runtime. When the schema changes, Minimal adapts automatically. There is nothing to update because there is nothing stored.
Access Control Built In
Governed by default, not bolted on
Every endpoint is governed from day one. Roles, personas, power tokens, multi-tenancy — access control is the runtime itself, not a middleware layer you add later and forget to update.
Multi-Database
One runtime, every database
PostgreSQL, MySQL, ClickHouse, MariaDB — one runtime, one control plane, one governance model. Switch databases without rewriting your API layer. The schema is the contract. Other Databases in Beta testing.
Point. Connect. Ship.
Point Minimal at any database. Production-safe RESTful APIs exist immediately — for every table, every relationship, every schema change. No route files. No handlers. No backend changes ever again.
× 16 tables · auto-infered · governed · zero maintenance
Built with one natural language question. Live data. No engineering needed.
A question asked should not survive the meeting unanswered.
Ask in plain English — connected to Minimal Insights, AI composes a full visual application: geo-maps, KPI cards, trend lines, breakdowns, filters. Live-connected to production data. Not last Tuesday's export.
Customers write their own apps. The best ones become your product.
Every customer-specific request that lands in your backlog is a ticket your team shouldn't be writing. Forge gives customers the ability to ask a question in plain English and get a live, governed application in return. No sprint. No wait. No generic version that half-fits everyone.
The old cycle
"The best customer-built apps graduate to product features. Customers write the roadmap. Engineering ships what matters, not what was loudest."
The Arc
The human did not know to ask. The system found the answer anyway.
Most patterns in production data are never seen. Not because the data is inaccessible — it is there. But because surfacing a pattern requires knowing to look for it, which requires having the question first.
Deep Thought explores schema and data autonomously. It surfaces anomalies, correlations, and patterns that no human thought to query for — presenting them as ready-to-use insights, scoped to the right team, at the right time.
Named after the computer in The Hitchhiker's Guide to the Galaxy. The character faced an answer without a question. Our Deep Thought finds questions you did not know to ask, and answers them with live data.
Market Signal
Three budgets.
One runtime addresses all of them.
The signal is already in the market.
of API teams still face collaboration blockers in production workflows — despite years of investment in tooling.
Postman · State of the API 2025
of developers are designing APIs specifically for AI agents — as 89% already use AI in their workflows. Infrastructure hasn't caught up.
Postman · State of the API 2025
of teams worry about unauthorised AI agent calls hitting production systems. Governance of AI-accessible APIs is the unsolved layer.
Postman · Security Report 2025
Embedded analytics and self-serve data access directly drive revenue and user engagement. The BI cost is a product competitiveness problem.
ThoughtSpot · Embedded Analytics
Minimal addresses all three budget lines simultaneously — API infrastructure, embedded BI, and AI tooling governance — from a single runtime. This is not a multi-product play. It is the same meta-logic engine applied to three problems the market has been treating as separate.
Philosophy
Meta-logic is not an architectural choice. It is the correct default.
A conventional program is a set of explicit instructions. It does exactly what it is told — no more, no less. Change the input structure and the instructions break. The program does not adapt. It fails.
A meta-logical system holds a model of structure and derives behaviour from that model at runtime. Change the schema and the system adapts, because the behaviour was never hard-coded in the first place.
Explicit instruction
I will query the users table, select id, name, and email, apply a filter where status = 'active', paginate with limit 10 offset 0, and return JSON.
Meta-logic
Get active users. Page 1. Fields: id, name, email.
The first requires a programmer. The second requires only intent. Minimal is the engine that bridges them.
"AI coding tools are a faster path to the same destination: a codebase full of routes nobody asked for. Minimal is the off-ramp."
Why this matters more now than before
Every AI coding tool — Copilot, Cursor, every LLM-based code generator — operates on the explicit instruction paradigm. It writes code. Code that must be reviewed, tested, maintained, and eventually deleted.
AI code generation makes the argument for meta-logic stronger. If the problem with backend boilerplate was that it was slow to write, AI has partially addressed that. But the problem was never speed of writing — it was existence.
Code generated automatically still exists. It still breaks on schema changes. It still requires maintenance. AI tools generate technical debt at scale rather than eliminating it.
The Prediction
Minimal is infrastructure for the slim tech team.
Every team we speak to is trying to ship with fewer engineers than the previous generation assumed you needed. Minimal is the answer — not because engineers are smarter, but because the mechanical translation layer no longer needs to be written at all.
"A two-person backend team doing what used to need ten."
Custom Definitions
Custom Definition is your logic layer.Deploy as an endpoint.
Write SQL, Starlark, or Expr — deploy instantly as a versioned, governed REST endpoint. No handler. No middleware. No build pipeline.
Activity Log
Complete visibility. Zero setup.
Every API call, every definition hit, every system event — live, audited, filterable. Method, path, latency, type, timestamp. Know exactly who is calling what, when, and how fast.
Filter by
GET · POST · PUT · DELETE · PATCH
Type
Auto · Custom · System
Time window
15m · 1h · 6h · 24h · 7d · Custom
Latency
Per-call tracking
Built for the people who move data.
Founders & Engineers
Ship your data product in hours, not months.
Stop building the same data exposure patterns for the fortieth customer. Auto API, Custom Definitions, and Forge let a two-person team do what used to need ten — without the boilerplate, without the maintenance, without the backlog.
Start building →Database Companies
The product layer your database has always needed.
Minimal is the API and product surface that turns any database into a complete data platform. Multi-tenant. Governed. AI-native. Multi-database. The infrastructure your customers need — already built.
Explore the runtime →Investors
Three budget lines. One runtime.
93% of API teams face production blockers. 24% are building for AI agents. 51% fear unauthorized calls. 60% need embedded analytics. Minimal is the single runtime that captures all three budgets — API infrastructure, BI, and AI tooling governance.
See market signal →The Builders
Built by enginners who felt the pain.
Harish S S
CEO & Co-founder
29 years building the systems enterprises run on. Senior Staff Engineer at VMware, Distinguished Engineer at Walmart Global Tech, Fellow at Cyware, Architect at HPE. Three US patents in workload assembly and OS-level runtime architecture. Inventor of Minimal's core inference engine — the schema-aware runtime that derives governed API behaviour at query time, with no code generation and no stored artefacts. The tax he is eliminating is one he watched teams pay at every company he led.
VMware · Walmart Global Tech · HPE · Cyware · 3 US Patents
LinkedIn →Gowthaman D
CTO & Co-founder
22 years building the observability and infrastructure layer of VMware's software-defined datacenter. Staff-2 Engineer. Designed Cloud Observability from scratch — 40MB JVM footprint, 8% vCPU utilisation — and led the SDDC instrumentation architecture across the full VMware stack. Core engineer on HPE OneView.
VMware Staff-2 · HPE OneView · SDDC Architecture
LinkedIn →"We are building the infrastructure layer the next decade of data products will run on — and having honest conversations with the people who see the same thing."
Littlebit Labs · hello@littlebit.in
Littlebit Labs
The tax is now, Zero.
Imagine, express, done.
An infrastructure layer you can talk to, literally.