HomeVision Tech Stack

AI-powered collateral underwriting for mortgage lenders and appraisers

Real Estate San Francisco, CA 11–50 employees Founded 2020 Privately Held

HomeVision built MIRA, a document-understanding platform that automates collateral review across appraisals, photos, floor plans, and loan files. The tech stack—Go, Python, TensorFlow, PyTorch, PostgreSQL, Redshift, dbt—reflects a mature ML infrastructure focused on computer vision and NLP. Engineering-heavy hiring (5 engineers, 1 data) with mid-to-senior seniority distribution suggests they're scaling core ML capabilities while their active projects prioritize expanding MIRA's loan-file coverage and productionizing valuation models.

Tech Stack 24 technologies

Core StackGo PostgreSQL React TypeScript Java C++ Python AWS Elasticsearch Docker NumPy scikit-learn TensorFlow PyTorch Terraform dbt Redshift SciPy Jupyter SQL Streamlit Git

What HomeVision Is Building

◆Challenges

Review time reduction
Covering entire loan file
Expanding mira to cover entire loan file
Automating income assets credit analysis
Labor-intensive document processing
Manual workflow processes
Inefficient onboarding process
Scaling infrastructure for saas products
Maintaining reliability and observability
Document review quality and efficiency

▲Active Projects

Building interfaces and infrastructure for mira
Integrating ai services into core saas platform
Frontend data visualizations
Expanding mira to cover entire loan file
Workflow automation products
Productionizing a valuation model
Saas infrastructure on aws
Developer tooling platform
Authentication and reliability projects
Dbt model maintenance

Hiring Activity

Decelerating8 roles · 1 in 30d

Department

Engineering

Data

Executive

Product

Seniority

Mid

Senior

Junior

Staff

Notable leadership hires: Chief of Staff

Company intelligence

Find more companies like HomeVision by tech stack, pain points and active projects

Get started free

About HomeVision

HomeVision delivers automated collateral underwriting for lenders, appraisal management companies, and mortgage professionals. The MIRA platform integrates with popular LOS and AMS systems to examine appraisal documents, photographs, maps, floor plans, and narrative content, flagging inconsistencies and risk factors that would otherwise require manual cross-referencing. Founded in 2020 and headquartered in San Francisco, the company operates a small, focused team primarily based in the United States with some presence in Argentina. The product roadmap centers on expanding coverage across entire loan files and automating income, assets, and credit analysis workflows.

HeadquartersSan Francisco, CA

Company Size11–50 employees

Founded2020

Hiring MarketsUnited States, Argentina

Frequently Asked Questions

What technology does HomeVision use for document analysis?

HomeVision's MIRA platform is built on Go, Python, TensorFlow, PyTorch, and PostgreSQL, with Redshift for analytics, Elasticsearch for search, and React for the frontend. The stack emphasizes computer vision and NLP capabilities.

Is HomeVision hiring engineers?

Yes. HomeVision has 5 active engineering roles posted, with mid-to-senior seniority levels. They are also hiring for data, product, and executive positions, with 8 total open roles across the company.

What is HomeVision working on?

Current projects include expanding MIRA to cover entire loan files, productionizing a valuation model, integrating AI services into the core SaaS platform, and building workflow automation products for loan processing.