Milena Tanui — AI Agent Builder & Automation Architect

About

Low-code. High impact.
Built with intention.

I'm Milena — an AI Agent Builder and Automation Architect who specializes in a low-code approach. I use prompt engineering, CrewAI, and AI-assisted development tools like Replit and Lovable to build intelligent, production-ready systems fast.

My background includes leading a team of AI trainers at Scale AI — one of the world's leading AI companies — giving me a rare combination of hands-on building skills and a deep understanding of how AI systems learn and perform in the real world.

I believe prompt engineering is a superpower. You don't need to write every line of code to build something powerful — you need to know what to build and how to make it think.

🎓

AI for Everyone

DeepLearning.AI / Coursera

🎓

Elements of AI for Business

MinnaLearn / Reaktor

Stack

🤖 AI Agent Building

CrewAI Claude Prompt Engineering Replit Lovable

⚡ Workflow Automation

Make Zapier n8n API Integration

Projects

What I've built.

Low-Code · AI Agent

GradeFlow

An agentic AI tool that ingests any course content — raw text, learning objectives, or module summaries — and instantly generates a fully structured, difficulty-calibrated assessment. No LMS required. Just a shareable link your team can access instantly.

CrewAI Replit Claude Multi-Agent L&D EdTech

View on GitHub →

Demo Video

Screenshots

01 — Input

Course content entry interface

02 — Output

AI-generated structured assessment

03 — Workflow

Multi-agent orchestration in CrewAI

Agent Workflow

01 Course content input (text / objectives / summaries)

↓

02 Agent 1: Content Analyser — reads & maps learning outcomes

↓

03 Agent 2: Assessment Builder — generates calibrated questions

↓

✓ Structured assessment + shareable link delivered

Low-Code · Automation

Invoice Automation Pipeline

Built to explore how AI and automation tools can eliminate manual invoice processing. The pipeline watches Gmail for PDF attachments, uses Claude AI to extract structured invoice data, routes low-confidence extractions for human review, and delivers a weekly HTML summary email — all without human intervention.

n8n Make Claude API Google Sheets Gmail JavaScript AI Extraction Prompt Engineering

Demo Video

Workflow Screenshots

01 — Canvas

n8n pipeline all nodes green

02 — Extraction

Claude AI extracting invoice JSON

03 — Sheets

Auto-populated vendor data

04 — Digest

Weekly summary email to owner

05 — Digest Workflow

Scheduled n8n digest pipeline

Pipeline Workflow

01 Gmail trigger detects invoice email with PDF attachment

↓

02 PDF converted to base64 via JavaScript code node

↓

03 Claude AI extracts vendor, invoice number, amounts, dates

↓

04 Confidence router — clean extractions vs needs review

↓

05 Google Sheets updated automatically by vendor

↓

✓ Weekly HTML digest emailed to business owner

No-Code · Automation

Team Issue Tracker

An end-to-end issue reporting automation built on Zapier. When a team member encounters a problem, a form submission triggers an automated chain — logging to Google Sheets, sending a confirmation email to the reporter, and firing a Slack alert to the responsible stakeholder for immediate action.

Zapier Google Sheets Slack Email Automation No-Code

View on GitHub →

Demo Video

Workflow Screenshots

01 — Formatter

Text transform step setup

02 — Full Flow

Google Forms → Sheets → Gmail → Slack

03 — Dashboard

Live published Zaps overview

Coming Soon

Next Project

More intelligent agents and automation systems in development. Follow on GitHub to get notified when new projects drop.

Follow on GitHub →

How I Build

From idea to
working agent.

Define the Logic

I work with Claude to articulate the workflow, design the agent structure, and generate structured prompts that will drive the system.

Tool: Claude

Orchestrate the Agents

The structured prompts are used in CrewAI to define and orchestrate multi-agent workflows — mapping how agents collaborate to complete tasks.

Tool: CrewAI

Build the Interface

Replit or Lovable is used to rapidly build and style the product interface, with AI-assisted development accelerating the entire process.

Tool: Replit / Lovable

Connect & Deploy

The frontend and agents are connected via API, tested end-to-end, and deployed. Automation layers are added with Make, Zapier, or n8n as needed.

Tool: API + Make / n8n

Milena
Tanui AI Agent Builder & Automation Architect

Low-code. High impact.
Built with intention.