the unconcealed.

Independent AI research & engineering

Building infrastructure for intelligent systems.

We build and test the systems around capable AI: agent coordination, tool execution, oversight, and efficient inference.

Explore our work Our approach

Independent · India · Founded August 2026 · Active development

The surrounding systemFIG. 01
Human authority
ModelsTools / workers
The right next step

Route between models, delegate to agents, and make coordination explicit.

Select a layer to explore Conceptual architecture

Beyond the model

Agent infrastructure Developer tooling Observable execution

01 / Our work

Work in progress.
Systems taking shape.

Three connected areas of active research and product development. Our systems are being built and validated ahead of broader releases.

01 / In research & development
FIELD NOTE / 01ILLUSTRATIVE
Handoff schematic · Not a product screenshot

Agent infrastructure

Exploring model routing, task delegation, and tool use for complex AI-assisted workflows.

Delegation · Routing · Context
02 / In research & development
FIELD NOTE / 02ILLUSTRATIVE
Trace schematic · Not a product screenshot

Observability & control

Exploring execution traces, agent visibility, evaluation, and governance around tool use.

Traces · Evaluation · Authority
03 / In research & development
FIELD NOTE / 03ILLUSTRATIVE
Resource-flow schematic · Not a product screenshot

Efficient AI systems

Investigating latency, memory use, and resource trade-offs on constrained hardware.

Inference · Latency · Resource use
Current stage / 2026

Independent, bootstrapped and in active research and product development. Our underlying systems are not yet broadly released.

Explore open questions

Proposed architecture / Concept

From intention
to evidence.

A model response is only one part of an agent workflow. We research how to coordinate tasks, define execution boundaries, and make outcomes easier to inspect.

Explore the open questions
PROPOSED FLOW / CONCEPT ONLYFIG. 02
  1. 01 / TASKDefine intentInputs · context
  2. 02 / ROUTESelect next stepModels · delegation
  3. 03 / EXECUTEWork with boundsTools · authority
  4. 04 / EVIDENCEDesign for evidenceResults · uncertainty
Concept illustration — not a live product interface.

02 / How we think

Capability needs
control.

Better models expand what is possible. Better systems make their work inspectable and dependable.

We design around orchestration, permissions, context, cost, and verification—not just model capability.

P.01

Evidence over assumptions

Expose what happened, why it happened, and what still needs to be verified.

P.02

Control by design

Make permissions, intervention, and human authority part of the architecture.

P.03

Efficiency matters

Use compute, context, and intelligence deliberately. Measure the tradeoffs.

P.04

Understandability matters

Build systems people can reason about, especially when things go wrong.

Development cycle / Guiding approach

How we approach
the work.

  1. 01

    Research

    Identify the constraints and questions.

  2. 02

    Build

    Turn an idea into a testable prototype.

  3. 03

    Validate

    Plan tests and record observable results.

  4. 04

    Refine

    Use results to guide the next design.

A description of our working method, not evidence of shipped products or measured performance.

03 / Open questions

Questions guiding
the next iteration.

Open engineering questions guiding our current research and development work. These are not released products.

01

Coordination & routing

Agents / Models

How should a system decide who does the next piece of work?

We’re exploring delegation, model selection, and coordination between agents. The goal is to make responsibilities and handoffs explicit, with useful boundaries on execution.

02

Context & memory

State / Continuity

What should a system carry forward, and what should it leave behind?

We’re researching how instructions, working state, and memory move across tasks. Relevant context should be available where it is needed without accumulating unnecessary cost or ambiguity.

03

Evidence & governance

Evaluation / Authority

Can a person understand, evaluate, and interrupt autonomous work?

We’re exploring execution traces, failure visibility, and evaluation alongside permissions and human authority. Observability and governance need to connect to the actions a system actually takes.

04

Efficient execution

Inference / Compute

How can capable AI become practical within limited resources?

We’re investigating inference, memory movement, latency, and resource use on constrained hardware. The focus is on understanding engineering tradeoffs before making performance claims.

Research directions, not released products.

04 / The initiative

Independent.
Founder-led.
Founded 2026.
Based in India.

An early-stage initiative.
Built with intention.

Founded in August 2026, The Unconcealed is an independent AI product and research initiative exploring the infrastructure needed when AI systems become capable enough to perform meaningful work autonomously.

We focus on building and validating systems around capable AI: agent orchestration, tool execution, observability, and practical efficiency.

Prathamesh EkadFounder, The Unconcealed

05 / Start a conversation

Build with clarity.
Start a conversation.

Working on the infrastructure around AI?
We’d like to hear what you’re thinking about.

pratham@theunconcealed.com