← nawal:~$ technologies

nawal:~$ technologies/ai-agents

AI agent development with Claude & GPT.

Agents that read emails, process documents, update your systems and draft reports — connected to your data and supervised so they stay useful. Automation that earns its place.

  • ★ 5.0 on Clutch
  • US-timezone senior team
  • You own the code

What we build with AI.

Practical agents for the busywork your team shouldn't be doing.

01

Inbox & document triage

Read incoming email, invoices and forms; classify them, extract the data and route them to the right place.

02

Back-office automation

Reconciling invoices, updating the CRM, chasing missing data — the repetitive steps between your systems.

03

Reporting agents

Weekly reports drafted from your real data, ready for a human to review and send.

04

Assistants over your own data

Ask questions about your operation in plain language and get answers grounded in your documents and database.

05

Workflow automation with n8n

Connect your tools into automated workflows, with AI steps only where they add something.

06

Evaluation & guardrails

Test sets, approval steps and audit logs, so you know how well an agent performs before you trust it.

nawal:~$ why-ai-agents

Why agents — and how we keep them honest.

  1. 01

    Hours back every week.

    The repetitive work moves to software, and your team keeps the judgment calls.

  2. 02

    Built into your systems.

    Agents read and write to the tools you already use — not a separate chat window nobody opens.

  3. 03

    Humans stay in the loop.

    Anything that matters gets a review step. You decide what the agent can do on its own.

nawal:~$ tools

The AI agents toolbox.

Models
Claude · GPT
Orchestration
n8n · Elixir services
Data
PostgreSQL · Embeddings · Your documents
Integrations
Email · CRM · Slack · Webhooks
Safety
Evaluations · Approval steps · Audit logs

nawal:~$ engagement

Ways to work with us.

Need an engineering department, not just developers?

Fractional CTO + dedicated team

A CTO who joins your leadership conversations and a senior squad that builds. One point of accountability for architecture, timelines and delivery.

Have a defined system to build?

Fixed-price project

A free one-week discovery, then a proposal with scope, timeline and a closed price per phase. Demo every Friday; the code is yours.

Already live and need it kept healthy?

Ongoing support & evolution

Monthly support for what's in production — upgrades, monitoring, performance and the next features. Optional, never a condition.

nawal:~$ process

How we work.

Four phases. The first one is free — not as a discount, but because we won't quote a system we don't understand yet.

01

Discovery

We sit with the people who do the work, not just the person who called us. We map how the operation actually runs — including the workarounds nobody documented. You keep the map either way.

1 week · free

02

Proposal

Scope, sequence, timeline, closed price — written so you could hand it to someone else and they'd understand it. You sign it or you don't.

1 week · fixed

03

Build

Two-week sprints. A working demo every Friday, run by the engineer who built it. Slack and repo access from day one — you can watch commits land.

8-24 weeks

04

Handover

We train your team on the system they now own. Monthly support if you want it, never as a condition. The code works without us.

ongoing · optional

nawal:~$ questions

AI agents, answered.

Is our data safe?

We use business API tiers whose providers don't train on your data, keep sensitive data in your own systems, and send models only what a task needs. We'll walk you through exactly where data goes.

Will this replace our team?

It replaces tasks, not people. The goal is to take the repetitive work off your team so they spend their time on the parts that need judgment.

How do we know it actually works?

We measure it. Before an agent goes live we test it against real examples from your operation, and we keep a human review step until the numbers say it's ready.

What does it cost to run?

Model usage is usually small next to the hours saved. We estimate running costs in the proposal so there are no surprises.

Got something that deserves to exist?

Tell us what you're trying to solve. Within 24 hours you'll have a call booked and a first read on it — no cost, no commitment, no fine print.