Shehab Ismail
LinkedIn (opens in a new tab)

Navigation

AI / SYSTEMS

I build technology that survives contact with the real world.

Founder, CTO, and hands-on technology executive working across production AI, embedded systems, computer vision, cloud platforms, and interactive technology.

Explore the work
Conceptual operational trace from signal through firmware, edge, cloud, AI, human decision, and physical outcome. It is not captured data.

01 / PHILOSOPHY

Portrait of Shehab Ismail.

I tend to work where the layers meet.

AI models are useful until they have to safely take actions.

Cloud systems are easy until the network disappears.

Hardware is straightforward until thousands of people hit it every day.

Research is promising until the experimental setup has to leave the lab.

I’ve spent more than a decade working across those boundaries — as an engineer, researcher, founder, CEO, and CTO.

Start with the problem. Go as deep into the system as necessary.

02 / SELECTED WORK

THE THINGS I'VE BUILT.

Different industries. Same requirement: They have to work.

01 / HAPPINEST.AI

2024—NOW

CO-FOUNDER & CTO

Building AI that can actually take action.

HappiNest product and safety layers
HappiNest official full logo lockup.
  1. 01Interface inputMobile leasing workspace
  2. 02Safety / state / validationRequired before action
  3. 03Operational systemsControlled action path
HOMES LIVE
3,277+
STATES
11
↑ LEAD → LEASE
39%Dean & DeWitt
HOMES CONTRACTED
6,500+and growing

THE PROBLEM

AI answering a question is easy. Safely taking actions inside operational systems is much harder.

MY ROLE

  • AI architecture
  • Engineering
  • Product
  • Infrastructure
  • Integrations
  • Reliability
HappiNest operational flow
  1. Stage 1
    • Lead
  2. Stage 2
    • Conversation
  3. Stage 3
    • AI
  4. Stage 4
    • Safety / state / validation
  5. Stage 5
    • Property operations
  • Lead leads to Conversation
  • Conversation leads to AI
  • AI leads to Safety / state / validation
  • Safety / state / validation leads to Property operations

The model decides what should happen. The surrounding system decides whether it is safe for it to happen.

READ THE CASE STUDY

02 / THEME-PARK INTERACTIVES

INNOVATED STUDIOS

Work completed as part of my role at Innovated Studios.

Conceptual interaction schematic showing sensor input, embedded processing, game state, and physical response.

CONCEPT STUDY

Firmware meets physical experience.

LED DEVELOPMENT
Rambi’s Bongo Breakout
FIRMWARE
Accelerometer-driven interactives
DEPLOYMENT
Orlando / Hollywood / Japan

A physical interactive has a guest standing in front of it. There is no refresh button.

  1. 01 /ACCELEROMETER
  2. 02 /IMPACT DETECTION
  3. 03 /EVENT OUTPUT
READ THE CASE STUDY

03 / AXIS PORT

INNOVATED STUDIOS

Work completed as part of my role at Innovated Studios.

Making physical locations programmable.

AXIS Port rugged connected reader shown as an isolated product cutout.
  • 01RFID / NFC / BLE
  • 025G / WI-FI / ETHERNET
  • 03GNSS
  • 04REMOTE TELEMETRY
  • 05IP65
  • 0624+ HR ACTIVE RUNTIME

PERSONALLY BUILT / LED

I wrote the firmware.

I also led the product architecture and the hybrid edge/cloud work around it.

  • Firmware
  • Product architecture
  • Hybrid edge/cloud architecture
  • Engineering leadership

THEME-PARK ENVIRONMENTS / ORLANDO + CALIFORNIA

READ THE CASE STUDY

04 / MENTAL EDGE

2024—NOW

Turning video into structured human signals.

Mental Edge official logo.

PUBLIC CAPABILITY SURFACE / NO SUBJECT DATA

  • 01FACE DETECTION
  • 02EMBEDDINGS
  • 03RECOGNITION
  • 04VIDEO
  • 05FACIAL ANALYTICS
  • 06API

I personally trained the emotion-analysis model powering the current API.

What it returns is facial analytics from video — the face as it moves, not a claim about someone’s inner state.

VIEW CASE STUDY

CURRENTLY BUILDING

Formward

Your health plan shouldn’t be static when your life isn’t.

Built first to solve my own problem. Formward is an experiment in turning health data into decisions that continuously adapt to real life.

PRIVATE BETA / TESTFLIGHT

05 / RESEARCH

Conceptual diagram showing an unscaled illustrative seismocardiographic waveform beside a simplified chest sensor field with 32 sensor positions and four accelerometers. It is not captured research data or an exact experimental layout.

ILLUSTRATION / CONCEPTUAL EXPLANATION / NOT CAPTURED RESEARCH DATA

WHERE IT STARTED.

Spatial Distribution of Seismocardiographic Signals

Springer, 2021

DOI 10.1007/978-3-030-67494-6_5 (opens in a new tab)
  • EXPERIMENT DESIGN
  • HUMAN-SUBJECT TESTING
  • SIGNAL-ACQUISITION SOFTWARE
  • INSTRUMENTATION
  • PORTABLE SYSTEM DESIGN

A technically valid system and a practically usable system are two very different things.

VIEW RESEARCH

06 / SPEAKING

Shehab Ismail speaking on a panel hosted by the Harvard Alumni Club of Tampa Bay and the Penn/Wharton Club of Greater Tampa.

Speaking

The Impact of AI & Robotics on Our Work-Optional Future

Panelist

Harvard Alumni Club of Tampa Bay × Penn/Wharton Club of Greater Tampa

August 2024

The two alumni clubs ran it together. I was on the panel.

VIEW EVENT (opens in a new tab)

07 / ABOUT

Shehab Ismail outdoors.

About me

I’m a founder and CTO. I still write the code, the firmware, and the models when that’s what the problem needs. I started in biomedical signal-processing research at the University of Central Florida, then followed the work into software, telecommunications, embedded systems, AI, computer vision, interactive entertainment, and company building.

MORE ABOUT ME

LET’S TALK.

If you’re building something that has to work in the real world, I want to hear about it.

LinkedIn (opens in a new tab)Email shehab.projects@gmail.com (opens email client)