Immersely Logo

Know what to grow. Build what comes next.

AI emotion tech for game studios of tomorrow

Immersely helps game studios manage the AI transition with a suite of AI tools. At its core is JUDE, the Game DNA knowledge layer: it sees, thinks, plays and feels a game like a player, so studios know what to grow and can build tools to create what comes next.

Open Developer PortalTalk to Immersely

52 in-game KPIs · moment-to-moment · peak-end · See the full measure list

No SDK. No lab. Players play from home.

A playtest JUDE can read like a player.

Share a build. Players play at home. You get what to grow, and what to create next. Nothing to integrate, nothing to change in the game.

  1. Step 1

    Share your build

    Upload a build, or send a Steam key or store link. No SDK, no code changes, no engine plugins.

  2. Step 2

    Players play at home

    Your own community, or our panel, plays in their normal setup. Webcam on, their own smartwatch on, gameplay recorded.

  3. Step 3

    JUDE reads the session

    Emotion, behaviour, and the frame land on one timeline. JUDE sees, thinks, plays, and feels that session the way a player did. The report comes back with the clips that decide what to grow and what to build next.

What reaches JUDE

SmartwatchHeart rate and strain from the watch they already wear.
WebcamExpression and attention.
GameplayThe frame on screen.
One timelineAffect is never read without the moment it belonged to.

60 in-game KPIs

How JUDE reads a session

  • See

    Usability

    Legibility and efficiency. Goals, mechanics, and results are readable without searching.

    14 KPIs
  • Think

    Mental model

    Cognitive state. Goal, control, and competence. Cognitive, not emotional.

    3 KPIs
  • Play

    UX and game loop

    Hook, reward, mastery, friction, and momentum, minute by minute.

    18 KPIs
  • Feel

    Affect

    Webcam-inferred energy, mood, and tension. Always read with the on-screen frame, never alone.

    17 KPIs
  • Remember

    Moment vs memory

    A session is judged by its peak and its end, not its average.

    8 KPIs
Remember

What will they remember?

A session is judged by its peak and its end, not its average. Duration barely counts.

Session intensity chart comparing the peak, the end, and the average.

In-session gameplay

One session, three moments

Gameplay only. No webcam, no face, no player name.

First-person arena shooter: an armoured opponent jumps mid-air across a hazard-striped arena as the player's rifle lines up the shot.
Wikimedia Commons, Xonotic 0.8.2, GPLv2+. Stand-in action scene, not an Immersely session.
  1. See

    Read the fight

    The arena, the threat, and the next move are all on screen.

  2. Play

    The jump is the loop

    Mid-air engage. Is the fight still worth taking?

  3. Feel

    The hit is the peak

    The shot is about to land. The beat the session will remember.

What lands on the desk

A session rec. A content brief.

Two outputs from JUDE’s reading of the session — not a second how-it-works.

Diagnosis sheet: Keep/Fix rec from a read session. Amber telemetry, not a Delight/Friction caption.

Game Growth Recommendation

What to change in the live game.

Production brief: landed clip → encounter spec. Magenta studio board, not a 1-2-3 pipeline.

Game Content Creation

The next encounter, built from the beat that landed.

  • Unity
  • Unreal Engine
  • Steam
  • Epic Games
  • PlayStation

Founder-led, science-backed

The UK’s brightest tech minds in AI & gaming

Building a hyper-fidelity foundation model for player emotion

Immersely began with founder Goodness Akalazu, an electrical engineer and roboticist asking how a machine can read the person in front of it, in real time. He built the team around that question, bringing in a Cambridge-trained physicist and AI entrepreneur and an award-winning data scientist. Since 2022 we have pointed all of it at one goal: helping studios understand their players.

  • Portrait of Goodness Akalazu

    Goodness Akalazu

    CEO & Founder

    Electrical engineer and roboticist turned AI researcher. He built Immersely around real-time emotion detection, the same skill a robot needs to work alongside people.

    • Robotics and mechatronics
    • Electrical and electronics engineering
    • Entrepreneur in Residence, Kingston University
  • Portrait of Dr. Luke Robinson

    Dr. Luke Robinson

    Chief Science Officer

    Physicist and serial AI entrepreneur. He founded Hazy, winner of a $1M Microsoft AI Prize, and brings the science behind JUDE.

    • Physics PhD, Cambridge
    • Research Fellow, Oxford
    • Partner, Post Urban Ventures
    • De Montfort Medal
  • Portrait of Siddarth Deshpande

    Siddarth Deshpande

    Lead Data Scientist

    Award-winning data scientist working across materials engineering, biochemistry, healthcare and AI. He turns raw player signals into models studios can trust.

    • PhD Ambassador, NUS Singapore
    • Alibaba Cloud AI Innovation Award
    • bitsCrunch Datathon winner

Advisors

Kingston University BRAIN Labs

Behavioural Research Analytics In Neurotechnological Systems

  • 3 Advanced Degrees
  • 7 AI Startups
  • 160+ Researchers

Ready to work with JUDE?

Talk to ImmerselyOpen Developer Portal