We are an early-stage technology startup building a next-generation AI platform. This is one of several projects we are developing, and we are looking to build long-term relationships with the right people. We are a small team, and we price and engage accordingly — but for the right candidates, there is significant ongoing work ahead.
Project Overview
We are building a real-time emotional intelligence system that uses a standard video feed — think webcam or Zoom-style camera — to analyze the emotional and cognitive state of a user during live interactions. The system observes the person on camera and provides a live awareness layer that can be consumed by an AI system, platform operator, or application.
This is not a research project. We need a working system built by someone who has done this before.
What We Are Building
The system must be capable of:
- Micro-expression detection — identifying involuntary facial movements that occur in fractions of a second, mapped to core emotional states such as anger, surprise, disgust, contempt, fear, sadness, and happiness
- Macro-expression and mood tracking — sustained emotional states including boredom, frustration, enthusiasm, and disengagement
- Gaze and attention analysis — whether the user is focused, distracted, or looking away
- Engagement and interest scoring — a real-time signal of how engaged the user is with the interaction
- Cognitive load indicators — signs of confusion, concentration, or mental overload
- Deception and authenticity signals — stress indicators, inconsistency, and evasion cues (probabilistic, not deterministic)
- Arousal and valence mapping — real-time placement on the emotional arousal and valence plane
The output of this system feeds into an AI layer or platform operator dashboard as a live data stream. Latency must be low enough for real-time use, with a target of sub-100ms.
Privacy and Ethics
Because this system processes live camera data, candidates must address how raw video is handled, whether it is processed on-device or transmitted, and how the system is designed to protect user privacy. We take this seriously and expect candidates to as well. Candidates should also be transparent about the limitations and appropriate use boundaries of emotional detection technology.
Who We Are Looking For
You must have hands-on experience building systems in this space. Relevant backgrounds include:
- Affective computing or emotion recognition systems
- Real-time computer vision and facial analysis
- Facial action coding systems (FACS) or action unit (AU) extraction
- Behavioral analysis or human state detection
- Psychology-informed ML or human factors engineering
- Research or commercial work in lie detection, attention tracking, or cognitive state monitoring
Experience with tools such as OpenFace, MediaPipe, DeepFace, Azure Face API, or comparable frameworks is a strong signal. If you have built something in this space — even in a research context — we want to hear from you.
How We Select Candidates
We receive a high volume of applications. To find the right person, we will select a small number of finalists based entirely on the quality and specificity of your proposal. We will not conduct screening calls in place of a written proposal. Candidates who submit generic or non-specific responses will not be advanced.
To be considered, your application must include:
- A specific technical proposal — the exact technologies, models, and frameworks you would use, your reasons for each choice, and the tradeoffs you considered. General recommendations without specifics will not be reviewed.
- Your implementation plan — a phased approach with key milestones and your estimated timeline to deliver a working system that meets our requirements.
- Your proposed price — please provide a fixed project price or a detailed breakdown. We are a startup and expect startup-appropriate pricing. Candidates should factor this into their proposal honestly.
- Your estimated delivery date — when you can start and when you commit to delivering a functional system.
- Why you — describe the specific systems you have built that are relevant, your exact role in those projects, and what makes you the right fit for this engagement.
- Evidence of past work — links to demos, papers, repositories, or case studies. If your work is under NDA, describe the scope and scale without disclosing confidential details.
- References — 2 to 3 professional references from comparable projects, with name, relationship, and contact method.
- Full-time salary expectation — in addition to your project quote, please include what you would expect for a full-time ongoing role, as we have multiple projects in development and are looking for long-term collaborators.
Payment Terms
Payment will be made upon delivery of a working system that meets our specifications and requirements. We are a fair partner and will work collaboratively throughout the engagement, but final payment is tied to a deliverable that performs to our satisfaction. We want this to be a successful engagement for both sides and will communicate clearly throughout the process so there are no surprises at delivery.
What Comes Next
We will review all proposals and select candidates for interviews based on how well their background matches our needs and how completely they answered the questions above. Partial or vague submissions will not move forward. Finalists will have a focused technical conversation before we make our selection.
This is one of several projects. The right person will have the opportunity to grow with us.
...
Show more