AI Engineer & Full-Stack Developer

0%

Skip to content

AI Engineer & Full-Stack Developer

Francesco Iaforte

I design and ship production software systems and applied AI solutions.

From architecture to deployment.

AI Engineer & Full-Stack Developer

Scroll

Introduction

Profile.

AI Engineer and Full-Stack Developer. I design and ship production software and applied AI. I work on management platforms, SaaS, web apps, data layers, APIs and automations, from architecture through to release. On the AI side I develop RAG and knowledge-base systems, agents with tool calling and multi-agent orchestration, fine-tuning pipelines, and ML models for forecasting and predictive analytics.

Training in management engineering helps me read processes, operational constraints and business goals before I settle on the technical approach.

AI Systems & Agents

RAG and knowledge bases · AI agents · multi-agent orchestration · fine-tuning · hybrid retrieval · document parsing · AI in operational workflows

Frontend & Mobile

Web applications · dashboards and management platforms · responsive interfaces · iOS and Android apps · embedded software

Backend & Application

APIs and backend services · business logic · authentication and roles · integrations · jobs and async processes

Data & Databases

Data modeling · relational and document databases · row-level security · ORM · vector databases · migrations and data layer

Cloud, Infrastructure & Delivery

Deploy · Linux environments · containers · CI/CD · server management · backups · SSL

Cross-Functional Skills

  • Requirements and feasibility analysis
  • Client and stakeholder communication
  • Technical-functional documentation
  • Problem solving
  • Priority management
  • Team working
  • Autonomy
  • Project coordination

Built while running Nexus Trade. Useful alongside the technical profile.

Evidence

Numbers.

InnovEdge

  • 0+full-stack software projects shipped to production
  • 0+API integrations and external services

AI Engineering

  • 0models fine-tuned for specific tasks
  • 0AI agents built and integrated
  • 0RAG / LLM Wiki systems shipped

Nexus Trade

  • 0krevenue in the first year, across 11 markets
  • 0+own and client stores run in production
  • 0+orders managed across own and client stores
  • 0collaborators coordinated

Next Tournaments

  • 0+years running an eSports platform in production

Career

Experience.

From management software to digital commerce. The common thread is shipping systems to production and keeping them running.

Inventory

Technologies.

Languages, frameworks, platforms and tools used in professional and personal work, grouped by area.

  • Next.js
  • React
  • TypeScript
  • Node.js
  • Python
  • PostgreSQL
  • Docker
  • Supabase
  • PyTorch
  • Hugging Face
  • Shopify
  • Spring Boot

AI & Machine Learning

Core area. The technologies I use to build AI systems, RAG pipelines, agents and retrieval.

AI Systems & Agents

Practices

AI AgentsRAG PipelinesMulti-Agent OrchestrationPrompt EngineeringContext EngineeringTool CallingVoice Agents

Frameworks

LangChainLangGraph

Retrieval & Knowledge

Practices

Hybrid RetrievalEmbeddings

Knowledge base

LLM WikiObsidianGraphify

Infrastructure

Vector DB

Models & Training

Practices

Fine-Tuning

Techniques

LoRAQLoRA

Libraries

PyTorchHugging FaceUnsloth

ML & Analytics

ML ForecastingPredictive Analytics

AI Platforms & APIs

Cloud AI

Vertex AIAmazon Bedrock

Model APIs

OpenAIMistral

Tooling

Google AI Studio

Software Engineering

Core area. Frontend, backend, data and infrastructure. What I use to take software from first design to production.

Frontend & Mobile

Frameworks & libraries

ReactReact NativeNext.jsAngularFlutterTailwindPrimeNGCapacitorJS

Languages

TypeScriptJavaScriptHTMLCSS

Tooling

XcodeAndroid Studio

Backend & Application

Runtimes & frameworks

Node.jsNestJSLaravelSpring Boot

Languages

PythonPHPJavaC++

APIs, Integrations & Jobs

APIs & integrations

REST APIWebhook

Async & scheduling

Queue / WorkerBullMQCron

Data acquisition

Apify

Data & Databases

Databases

PostgreSQLRedisMySQLMariaDBMongoDBSQLite

ORM & platforms

SupabasePrismaEloquent

Formats

SQLJSONXMLYAML

Cloud, Infrastructure & Delivery

Cloud & infrastructure

Google Cloud Platform (GCP)DockerKubernetesOpenShiftLinux

Delivery

GitGitHubJenkinsSSHWHM / cPanelSSLBackupsServer migrationsMulti-environment deploy

AI-Assisted Engineering

Coding agents and automation are part of how I work. I use them for implementation, review, tests and repetitive tasks, and I keep control of architecture, technical choices and code.

Coding Agents & Automation

Coding agents

Claude CodeCursorCodexGoogle Antigravity

Automation

n8nWorkflow Automation

Complementary Experience

Tools used in production at Nexus Trade and on client work. They sit next to the engineering profile.

Commerce & Payments

Platforms

ShopifyWordPressMagentoPrestaShop

Payments

StripePayPalKlarnaRevenueCat

Growth & Analytics

GA4CROMeta AdsGoogle AdsTikTok AdsLinkedIn AdsKlaviyo (Email & SMS)

Content & Creative AI

CanvaCapCutHeyGenHiggsfieldRask.aiElevenLabs

Projects

Projects.

A selection of systems designed and shipped to production, from construction sites to hospitality. Every project is shown through its real interfaces.

AI Operations Hub — Prospecting, CRM & Voice Automation — Operations view: leads, calls, orders waiting for confirmation, open tasks and acquisition channels.
17 media

01

AI Operations Hub — Prospecting, CRM & Voice Automation

Production B2B operations platform

A production B2B operations layer combining prospecting, CRM, AI voice operations and request management. The system runs acquisition campaigns, enriches and qualifies leads from traceable sources, handles calls through a backend-connected voice agent, and turns conversations into order drafts, tasks, refund cases and activities that can be synchronised with external management systems.

#React#TypeScript#NestJS#PostgreSQL#Redis
DeepScout — Research flow connecting orchestration, planning, specialised agents, verification, synthesis and report generation.
4 media

02

DeepScout

Evidence-based multi-agent research

DeepScout is an open-source system for structured research grounded in sources and evidence. It breaks an objective into tasks, coordinates agents, uses hybrid retrieval, and measures coverage to investigate material gaps. It includes Human-in-the-Loop controls, persisted evaluations, and versioned adaptive policies with monitoring and rollback.

#Python#TypeScript#Next.js#FastAPI#PostgreSQL+pgvector
OptiBuild — Site dashboard: active sites, total activities, assigned operators, revenue and work-progress per job

03

OptiBuild

Intelligent construction site management

A platform that digitizes operational, technical and administrative construction site processes: costs, accounting, work progress statements, quantity takeoffs, documents, resources and dashboards. I worked on the architecture, data layer (170+ tables, RLS across 8 roles, audit trail), management modules, API integrations and AI parsing of quantity takeoffs up to 500+ pages long. Presented at Edil Expo 2026 in Rome.

#React#Vite#Laravel#PHP#CapacitorJS
Food, Retail & Logistics Management Platform — JARVIS, the platform's AI assistant: suggested commands and hands-free voice mode
19 media

04

Food, Retail & Logistics Management Platform

Restaurants, minimarkets and cash & carry

A platform for restaurants, minimarkets and cash & carry businesses. I worked on CRM, warehouse, supplier, cost, HACCP and loyalty modules, on logistics (driver tracking, intelligent route assignment) and on integrations with scales, POS and fiscal registers, plus what-if simulations on pricing, suppliers and new store locations.

#React#TypeScript#Vite#Tailwind#Supabase
TemaSuite — Dashboard: arrivals and departures, payments, pending invoices, missing check-ins, monthly revenue and latest bookings
3 media

05

TemaSuite

SaaS for property managers and hospitality

A multi-company management platform for bookings, guests, units, accounting and rates. I contributed to the full-stack architecture, backend API, frontend, data model and operational workflows, including the Channel Manager (pricing, availability, rate plans, OTA sync), the Price Advisor for competitor monitoring, and online check-in with OCR and identity verification, with data isolation, auditing, automated tests and controlled deployment.

#Angular#PrimeNG#Laravel#PHP#Stripe
Mielepiù — Operational dashboard with open requests, quotes sent and accepted, overdue tasks and a per-operator summary
2 media

06

Mielepiù

B2B platform with an AI agent

A platform for managing customers, orders, catalog, activities and KPIs for sales teams, with integrated Email and WhatsApp channels. I built management modules, workflows between requests, activities and operators, and AI components for request/task classification, customer/product clustering and predictive analytics on reorders and commercial KPIs.

#React#TypeScript#NestJS#Prisma#TanStackQuery
ANCC — Inspection pro forma: lines per installation type, statutory charges, totals summary and payment status

07

ANCC

Technical inspections and the administrative cycle

A management system for a technical inspection body carrying out periodic checks on installations. It covers the whole chain: records for leads, customers, sites, installations and technicians, inspection assignments, scheduling, inspection execution, document archive and administration, with linked invoicing and renewals.

#Laravel#PHP#React#TypeScript#Vite
WIP — Warehouse dashboard: products, jobs, available and committed quantities, conditions and items to restore

08

WIP

Event warehouse management

A warehouse management system for event set-ups, as the product describes itself. It holds together the product catalog, furnishings owned by the customer, the share committed to jobs, the restoration cycle and delivery notes, with master data and user roles.

#Laravel#PHP#React#TypeScript#Vite
Athleta — Mobile dashboard: active goal, weight, macros for the day and the trend over time
4 media

09

Athleta

AI fitness app with per-user memory

A personal web app for training, nutrition and progress tracking, with an integrated AI agent grounded in application context, memory and a dedicated knowledge base; suggestions are applied only after user confirmation. I handled the full-stack architecture, database modeling, mobile-first UI, AI integration, data validation, deployment and RAG/LLM Wiki systems over data and documentation.

#React#TypeScript#Vite#Supabase#PostgreSQL
MVP Athletic — Trainer area: customisable check-in questionnaires on stress, nutrition and training

10

MVP Athletic

Fitness platform on web, iOS and Android

A fitness platform available as a web app and as iOS and Android apps, with two ways to use it. A professional registers as a Personal Trainer and manages their own clients, who sign in as that trainer's clients. Alternatively a user registers independently and manages their own gym programme without a trainer.

#Laravel#PHP#React#TypeScript#Vite
IAF Agent Operating System — The Agent OS LLM Wiki explored as a graph: policies, agent matrix and the branches of each supported project

11

IAF Agent Operating System

Control plane for AI-assisted development

An experimental personal project: an orchestration layer on top of Claude Code, Cursor, Codex and Antigravity, designed and built independently. Agents and subagents, skills, rules, hooks, multi-model routing, preflight checks, security policy and a centralized knowledge base. Goal: AI-assisted development that is controllable, verifiable and traceable.

#ClaudeCode#Cursor#Codex#GoogleAntigravity#LLMWiki
n8n Automations — Main workflow: one webhook endpoint per step of the sales cycle, from new case to invoice
2 media

12

n8n Automations

An orchestrated sales cycle, with AI in the middle

n8n automations that orchestrate an entire sales cycle, from a new case to the invoice. The main workflow exposes a webhook endpoint for every step; a dedicated sub-workflow uses an AI agent to produce a document from the material gathered during a call.

#n8n#Webhook#GoogleSheets#GoogleDrive#GoogleCalendar

Systems

IAF Agent Operating System.

How a development task moves through the IAF Agent OS control plane, from request to final verification.

01 / 07

Task. A development task enters the system: a bugfix, a feature, a refactor, a security review.

Athleta

AI fitness app with per-user memory

In Athleta the AI proposes changes. The user confirms before anything is applied.

1. AI proposes

2. Confirm

Project detail

Education

Education.

Management Engineering studies (Logistics and Production)

University of Naples Federico II

Logistics, production, operations management, process analysis and optimization.

Scientific High School Diploma

2013 – 2018

Liceo Scientifico “Luca Pacioli” — Sant'Anastasia (NA)

Languages

  • ItalianNative
  • EnglishB2

Certifications

  • Anthropic Education · Jul 2026

    Deploying Claude on AWS Bedrock: API patterns, tool use and RAG.

  • Anthropic Education · Jul 2026

    Integrating Claude on Vertex AI: prompt engineering, tool use, RAG pipelines and agentic workflows.

  • Anthropic Education · Jun 2026

    Building AI applications with the Anthropic API: tool use, RAG, embeddings, MCP and agentic workflows.

  • Anthropic Education · Apr 2026

    Claude Code in software development workflows: automation, MCP, hooks, SDK and AI-assisted code review.

  • 05Automation & Workflow Fundamentals · API Integrations · AI Workflows

    n8n — Automation & Workflow · 2026

    Workflow automation, API integrations and AI workflows on n8n.

Contact

Contact

For collaborations, AI/full-stack projects, or a technical conversation, reach me by email or LinkedIn.

Download CV

The message is delivered to my Gmail. I do not store the data in a database.