Hi, I'm

Antonio Consiglio

Computer Vision Engineer specializing in end-to-end AI systems — from training and deploying models on edge hardware to building Agentic workflows with LLMs and VLMs. Passionate about turning research into production-ready solutions for real-world impact.

4+
Years Exp.
3
Languages
Italian English Spanish
Antonio Consiglio

Areas of Expertise

Core competencies built through years of deploying real AI systems.

Computer Vision Layer

Training and optimizing deep learning models for object detection, image classification, and semantic segmentation on custom datasets.

PyTorch Object Detection Custom Training

Edge AI & Inference

Exporting to ONNX and converting to target-specific formats for high-efficiency deployment across heterogeneous edge devices.

TensorRT ONNX Nvidia Jetson

Agentic AI Workflows

Architecting robust agentic workflows using LLMs and VLMs to automate complex reasoning and dynamic decision-making in real-time pipelines.

LLMs & VLMs FastAPI PydanticAI
Python3
C++
PyTorch
TensorRT
ONNX
Docker
Linux
HuggingFace
FastAPI
GenAI / vLLM

Professional Journey

Real-world impact across robotics, autonomous systems, and enterprise scales.

February 2025 — Present

Applied Scientist – Computer Vision

Presight | G42 · Abu Dhabi, UAE

  • Design and deploy end-to-end Computer Vision & Agentic AI solutions for smart city and enterprise-scale applications.
  • Train and optimize deep learning models (detection, classification, segmentation); export to ONNX and convert for high-efficiency edge deployment.
  • Lead development of a new GenAI project from the ground up, driving latency improvements across legacy pipelines.
  • Architect agentic workflows using LLMs, VLMs, FastAPI and PydanticAI via real-time video analytics reasoning.
March 2023 — January 2025

Computer Vision Engineer

Lanit-Tercom Italia S.R.L. · Bari, Italy

  • Led and coordinated two R&D teams through the full project lifecycle for an international autonomous robotics client.
  • Researched, trained and deployed Semantic Segmentation & Object Detection models optimized for real-time edge performance via TensorRT.
  • Developed a C++ shared library with Python bindings, implementing path planning (ACO) into the robot's autonomous navigation stack.
November 2021 — February 2023

Robotics & AI Software Engineer

EPF Automation · Carrù, Italy

  • Developed CV applications for industrial automation using Python and C#, integrating deep learning detection into production.
  • Built a full perception-to-action pipeline managing Intel RealSense cameras, Cognex ViDi inference, and communicating coordinates to a Mitsubishi robotic arm through a custom PySide2 HMI.

Let's Connect

Interested in collaborating or discussing Agentic AI and Computer Vision? Send me a message below.

Or connect with me on professional networks:

LinkedIn GitHub