NeuroSmith · Capability statement
neurosmith.co · contact@neurosmith.co · Remote from Pakistan, working worldwide
NeuroSmith TechnologiesAI engineering for agencies
NeuroSmith is a small AI engineering team that works with agencies and studios as a white-label partner. When a client needs an AI system, we design and build it, and the agency delivers it under its own name. Each project has a written scope and a fixed project fee, so the agency can agree its own price with its client.A white-label AI engineering team for agencies and studios.
The model
A written scope, a fixed project fee, and the agency’s brand on the result.
The agency shares its client’s requirements, with an NDA first where the details are sensitive. We reply with an implementation plan and a fixed project fee for that scope.
The agency agrees its own price with its client. We don’t take part in that conversation.
We build to milestones and report to the agency. We can join client calls as part of the agency’s team, or stay in the background.
Every milestone comes with its source code, a demo video and an install guide, with documents prepared for the agency’s branding.
Scope changes are agreed in writing before work on them starts.
A fixed fee for an agreed scope, known before the agency quotes its client.
Nothing the client receives carries our name.
We don’t approach our partners’ clients.
An NDA before details, where they are sensitive.
Source code, a demo video and an install guide with every milestone.
A small team of named engineers: the four people below.
Work
Some of our work is public. Much is under NDA, as partner projects would be. NDA projects keep the engineering and drop the names.
In progressDelivered
Metrics appear only with a client’s permission.
Delivered
An AI music-learning app: photograph your lesson notes, get back a song about them.
A student photographs their notes. The app reads them, writes a summary pitched at the student’s level, has a language model write lyrics from it, and generates a song with karaoke lyrics.
What we built
in a Flutter app
In progress · substantially built
A real-time talking-avatar travel concierge you can speak to.
You speak to it in a Flutter app and it answers out loud, on screen, from a theme-park knowledge base built by hand from data sheets and reference files.
What we built
in a Flutter app
Delivered · Sep 2026
Agents that read an insurer’s estimate, find the missing line items, write the argument and keep negotiating.
What we built
It starts from the insurer’s estimate, a PDF, and works out what is missing: omitted items, and differences in quantity and scope. It writes the case for each one, then runs the negotiation by email. It sends the argument, replies safely to the answers and follows up on its own.
Disclosure: Client withheld by agreementMetrics: Shipped; metrics confidential.
Delivered · Sep 2026
Computer-vision quality checks of field inspections, from photos taken on site.
What we built
role-based, multi-tenant
The app guides the inspector through multi-view photo capture at each installation stage and checks that the photos were taken inside a geofence. A multimodal model judges them and returns a verdict with a confidence and a summary, and the system writes a branded PDF report.
Disclosure: Client withheld by agreementMetrics: Shipped; metrics confidential.
Phases 1–5 delivered
Ask the company’s documents a question; analytical answers come back with a chart they drew themselves.
What we built
role-based data rooms, shared chats
Retrieval-augmented generation over a company’s own files, for people who want answers rather than search results. Analytical questions come back with a chart drawn for that answer, titled in the language of the question.
Disclosure: Client withheld by agreementMetrics: Shipped; metrics confidential.
Status not stated
A newsroom of agents: crews that plan, write, edit, translate and publish.
What we built
plus a title-optimisation crew · auto-posting to WordPress
Content production as a pipeline of agent crews. A Planner, a Writer, an Editor and a Translator hand work along the line, and a separate crew optimises titles. Each agent runs on the model you choose, and finished pieces go out to WordPress automatically.
Disclosure: Client withheld by agreement
Want something like this built? Get in touch
Services
Eight areas we work in, in alphabetical order.
AI inside a product you sell: multi-tenant, role-based, built for paying customers rather than a demo.
Typical buildsA multi-tenant claims-negotiation platform; photo-based inspection QA with role-based access; document search with a separate index per tenant.
Assistants that answer from your own documents, and say so when the documents don’t cover the question.
Typical buildsAn assistant over company documents whose analytical answers arrive with a chart; a voice concierge that answers from a hand-built knowledge base (in progress).
Systems that judge what is in a photo, such as inspection checks.
Typical buildsInspection QA that turns guided photos into a verdict with a confidence and a report.
Neural networks designed and trained for the problem in front of you, when an off-the-shelf model is not enough.
Text, image, audio and video generation built into a product, with the plumbing that makes it repeatable.
Typical buildsPhotographed lesson notes turned into a song with karaoke lyrics.
LLM systems that read documents, reason over them and write, grounded in data you control.
Typical buildsAgents that read an insurer’s estimate, find the missing line items and argue for them; retrieval over company documents in five file formats.
Models that learn from your data to predict, rank or recommend, judged on data they have not seen.
Agents that carry a task through many steps, or a conversation, by voice, email or chat.
Typical buildsA negotiation agent that sends, replies and follows up by email on its own; a talking-avatar travel concierge (in progress).
Have a system in mind? Get in touch
How we work
Every project runs the same process and produces the same five things. If the details are sensitive, we sign an NDA before you share them. In a white-label project, “you” is the agency.
We sign an NDA first if the details are sensitive, read the brief and whatever data exists, and write the implementation plan.
You get
You decide
Scope, milestones, and what “done” means.
We design the system on paper (how data flows through it, the schema it lands in, every API it calls or exposes), then build towards the first milestone.
You get
You decide
Whether the architecture fits your systems before the build goes further.
We deliver milestone by milestone and fold your feedback in through revision rounds.
You get
You decide
What changes in each round.
We hand the system over with its documents, so your team holds everything the project produced.
You get
You decide
Sign-off.
Illustrative brief · not a client’s
“We want an AI chatbot that answers , , and is .”
Team
A small team of named engineers. These four are the people who build partner projects.
Chief Operations Officer
LinkedIn ↗ (profile of Safiullah Rahu, opens in a new tab) Website ↗ (personal site of Safiullah Rahu, opens in a new tab)
Lead LLM Engineer
Full-Stack Developer
AI Engineer
Standards
Every number on this site is a fact about us or our work.
That is why four of the six projects here are shown without client names.
If the details are sensitive, we sign an NDA before you share them.
Each one comes with its source code, a demo video and an install guide.
Nothing the client receives carries our name.
A model has two scores. The first is measured on the data it learned from, and it flatters. The second is measured on data it has never seen, and it tells you how the system will behave in your business. That is the number we work to.
We judge trained models on data they never saw.
When we train or tune a model, validation data is set aside first and stays aside.
We say so when AI is the wrong tool.
Sometimes a rule, a query or a form does the job better. We’ll tell you.
We don’t publish numbers we can’t show.
That is why this site has no results percentages: client metrics stay with clients.
Corrections: none yet. Last reviewed 8 October 2026.
Question not here? Ask it directly. Email your question
The agency agrees a fixed project fee with us for a written scope, and agrees its own price with its client. We build to milestones and report to the agency, which delivers the work under its own brand.
Per project, from the scope in the implementation plan. The fee is fixed for that scope. Scope changes are agreed in writing before work on them starts.
Only if the agency wants them to. Nothing the client receives carries our name, and we don’t contact our partners’ clients directly.
Yes, and before any details are shared. Much of our work is under NDA, which is why several projects here are shown without names.
Pakistan Standard Time, UTC+5.
Everything the project produced: the implementation plan; the architecture document (data flow, schema, API map); every milestone with its source code, a demo video and an install guide; and the hand-over documents.
Contact
Tell us a little about your agency and the project in mind. We’ll reply with an implementation plan and a fixed project fee.
We’ll tell you plainly if AI isn’t the right tool.
