Applied AI Software Engineering Intern - PropTech / FinTech (Unpaid)

KeyPath is an AI-enabled PropTech and FinTech platform designed to improve the economics of renting for both tenants and property owners.
For landlords, KeyPath uses data, tenant incentives, workflow automation, and analytics to help improve tenant retention, reduce turnover and maintenance costs, and increase property-level NOI.

For renters, KeyPath provides rewards, financial participation tools, and longer-term pathways toward wealth creation and homeownership.

KeyPath sits above landlords’ existing property-management systems rather than replacing them. Our technology integrates property, lease, payment, maintenance, renewal, and resident activity data into a common intelligence layer that powers our products.

The Role

KeyPath is seeking a technically strong Applied AI Software Engineering Intern to help build AI systems tied to real operating and financial outcomes in residential real estate.
This is not a beginner coding internship and not a prompt-engineering role.

You will work alongside KeyPath’s CEO, CTO and experienced software engineers, but you should already be capable of independently building, testing, debugging, and documenting software.

The objective is to build AI around:
proprietary workflow + proprietary data + measurable financial outcomes
rather than simply adding a chatbot to the KeyPath application.
AI Workstreams
Depending on experience and project priorities, you may work on one or more of the following:
Renewal and Churn Prediction
Build models that identify residents with an elevated probability of non-renewal and surface the behavioral factors contributing to that risk.
Reward Optimization
Develop recommendation logic that determines which retention action or incentive is most economically efficient for a particular resident or situation.

Example:
Resident renewal probability without intervention: 55%
Recommended intervention: $25 reward + targeted renewal outreach
Estimated renewal probability after intervention: 68%
Expected avoided turnover value: $900
Intervention cost: $25
Expected net owner value: $875

Maintenance-Risk Classification
Use work-order history, maintenance reports, document data, and other property signals to identify issues that may develop into larger maintenance expenses if not addressed.

Delinquency Intervention Intelligence
Identify payment-pattern changes and develop supportive intervention recommendations.

This functionality is intended to support communication and resident assistance. It will not be used to recommend eviction or other adverse housing actions.
Owner ROI Attribution

Build systems that connect KeyPath actions to landlord financial outcomes, including:
·       avoided tenant turnover
·       vacancy reduction
·       maintenance cost avoidance
·       incentive spending
·       renewal improvement
·       expected NOI contribution

Homeownership-Readiness Research

Explore educational indicators that help renters understand progress toward financial readiness for homeownership.
Any such system is informational only and may not be used to approve, deny, price, or condition access to housing or credit.
Example Technical Project

One current development project is the KeyPath Resident Retention and NOI Intelligence Engine.

The system consumes normalized resident activity such as:
Payment events
Lease expiration
Maintenance requests
Lease renewals
Move-in and move-out events
Reward activity
and returns:
Renewal risk
Maintenance risk
Recommended retention action
Confidence score
Reason codes
Expected landlord financial impact
An example output might be:
Renewal Risk: 71%
Recommended Action: Targeted renewal outreach + $25 reward
Model Confidence: 82%
Estimated Avoided Turnover Value: $865
Action Cost: $25
Expected Net Owner Value: $840

What You Will Do

You may:
·       Build Python services and AI/ML pipelines
·       Develop feature-engineering pipelines from resident and property events
·       Train and evaluate classification or prediction models
·       Build deterministic financial attribution models
·       Develop REST APIs using FastAPI or similar frameworks
·       Build synthetic datasets and testing environments
·       Implement model confidence and explainability
·       Create automated unit and integration tests
·       Build data-validation and model-evaluation tools
·       Integrate AI outputs into KeyPath APIs and dashboards
·       Create technical documentation and model cards
·       Participate in architecture and code reviews
·       Work through GitHub issues, branches, pull requests, and code reviews

Required Qualifications

Applicants should already have meaningful software-development experience.
You should have:
·       Strong Python programming skills
·       Experience using Git and GitHub
·       Experience building or consuming REST APIs
·       Understanding of data structures and software design fundamentals
·       Experience debugging your own code
·       Ability to write automated tests
·       Familiarity with SQL, MongoDB, PostgreSQL, or another database
·       Ability to work independently from technical requirements
·       Strong technical documentation skills
What You Will Gain
You will work on applied AI problems where the output must connect to a measurable business result.
You will gain experience with:
·       production-oriented AI architecture
·       PropTech and FinTech systems
·       behavioral event data
·       predictive analytics
·       AI explainability
·       financial attribution
·       API design
·       data engineering
·       real estate operating economics
·       startup product development
Strong work may become part of KeyPath’s production product roadmap.