Build the future of
real estate finance

This is not a typical ML engineer role. RealBooks is looking for someone who can build production AI systems AND understands how accounting actually works — debits, credits, general ledgers, chart of accounts, depreciation schedules, and tax classification logic.

You'll design and implement the machine learning models that power our Autonomous General Ledger and AI agent architecture (Penny, Dollar Bill, and Uncle Sam). Your models will classify transactions, extract data from receipts and invoices, optimize tax strategies, detect anomalies, and continuously learn from investor behavior — all within a system that must be accurate enough for tax filing and audit defense.

What you'll do:

• Design, train, and deploy ML models for financial transaction classification — mapping bank feed and credit card transactions to the correct GL accounts, properties, and IRS categories
• Build intelligent document processing pipelines for receipt and invoice data extraction (OCR, NLP, structured data extraction from PDFs, images, and email content)
• Develop tax optimization models that identify deductions, depreciation strategies, and cost segregation opportunities across real estate portfolios
• Implement anomaly detection for financial data — flagging misclassified transactions, duplicate entries, and reconciliation discrepancies
• Build and maintain the AI backbone for our three named agents: Penny (expense tracking), Dollar Bill (financial analysis), Uncle Sam (tax strategy)
• Collaborate with the engineering team to deploy models into production with proper monitoring, versioning, and feedback loops
• Work with the Product Owner to translate accounting domain requirements into ML system design
• Continuously improve model accuracy using investor corrections and real-world financial data

What we're looking for:

• 3+ years building production ML/AI systems — not just notebooks, but deployed, monitored, production models
• Strong foundation in NLP, document understanding, or financial ML
• Practical understanding of accounting fundamentals: double-entry bookkeeping, general ledger structure, chart of accounts, journal entries, trial balance, financial statements
• Experience with or strong interest in: real estate accounting, IRS tax categories (Schedule E, depreciation, capital gains), or property-level financial reporting
• Proficiency in Python, with experience in ML frameworks (PyTorch, TensorFlow, scikit-learn, or similar)
• Experience with LLM integration, prompt engineering, or agent architectures is a strong plus
• Familiarity with OCR and document processing tools (Tesseract, AWS Textract, Google Document AI, or similar)
• Comfortable working in a startup where you'll own the full ML lifecycle — from data collection to model deployment to monitoring
• Bonus: CPA, accounting coursework, or professional experience in bookkeeping or tax preparation
• Bonus: experience with real estate financial modeling or property management software

Compensation: Competitive salary + equity. This role is rare — the intersection of ML expertise and accounting domain knowledge — and compensation reflects that.

Location: Remote, with preference for candidates east of Denver (Central or Eastern time) to maximize collaboration with our US-based leadership and Moldova-based engineering team.

RealBooks is hiring a Full-Stack Engineer to build the platform investors trust with their entire financial life. You'll ship features across the stack — frontend, backend, data, and integrations — for a product that has to be polished enough to delight real estate investors and accurate enough to survive an IRS audit.

You'll work alongside our distributed engineering team — leadership in Sarasota, Florida and the core engineering team in Moldova. The work spans everything from the customer-facing web and mobile apps to the integrations that pull in bank feeds, parse receipts and HUD statements, and feed our Autonomous General Ledger.

What you'll do:

• Build customer-facing features across the web app (app.realbooks.io) and the mobile apps (iOS + Android) — TypeScript on the frontend, Python on the backend
• Ship integrations that ingest financial data from every angle: bank feeds, email forwarding (receipts@realbooks.io), receipt OCR, and HUD statement parsing
• Implement and maintain IRS schedule generation (Schedule D, E, F, 8949, 8582, 4562) with the correctness guarantees real estate investors and CPAs need
• Work on the systems that power our AI agents — Penny (expense tracking), Dollar Bill (financial analysis), and Uncle Sam (tax strategy) — building the data pipelines, classification feedback loops, and human-in-the-loop tooling
• Help maintain the multi-entity, multi-property data model that supports investors with complex legal structures (LLCs, trusts, S-Corps)
• Write clean, well-tested code for systems that handle financial data — correctness matters more than cleverness, and nothing ships without tests
• Collaborate across time zones with the engineering team in Moldova; participate in code reviews, design discussions, and shared roadmap planning

What we're looking for:

• 4+ years building production web applications, ideally across both frontend and backend
• Strong TypeScript and Python proficiency
• Experience with relational databases (PostgreSQL preferred) and the schema design that goes with multi-tenant, multi-entity SaaS
• Comfortable with cloud infrastructure (AWS, GCP, or similar) and modern deployment practices — CI/CD, infrastructure-as-code, observability
• Care about correctness — financial systems have no margin for "we'll fix it later"
• Comfortable working asynchronously with a distributed team across multiple time zones
• Bonus: experience with accounting systems, fintech, or real estate technology
• Bonus: experience integrating with banking APIs (Plaid, MX, Finicity) or document-processing services (AWS Textract, Google Document AI, OCR pipelines)
• Bonus: experience shipping AI/ML-powered features in a production setting
• Bonus: mobile experience (React Native, Flutter, or native iOS/Android)

Compensation: Competitive salary + equity. Early-stage hire with meaningful ownership in the company.

Location: Remote, with preference for candidates east of Denver (Central or Eastern time) to maximize overlap with both our US-based leadership and Moldova-based engineering team.

RealBooks is hiring an AI/Machine Learning Engineer to build the ML platform and agent infrastructure behind our three named AI agents — Penny (expense categorization), Dollar Bill (financial analysis), and Uncle Sam (tax strategy).

This is the systems-and-platform side of our AI work. You'll build the training pipelines, model serving, evaluation harnesses, and LLM/agent infrastructure that the team uses to ship intelligence into the product. You'll partner closely with our AI & Accounting Intelligence Engineer (who owns the domain-specific modeling) to make sure the platform is fast, observable, and continuously improving.

The work has to be accurate enough that a CPA can sign off on what comes out of it — millions of bank transactions classified, tens of thousands of receipts extracted, and deductions surfaced at scale.

What you'll do:

• Build and operate the ML platform: training pipelines, model registry, evaluation, deployment, monitoring, and a feedback loop that captures investor corrections and re-trains automatically
• Design and ship agent infrastructure for our LLM-powered features — prompt orchestration, tool use, RAG over financial data, and grounding/safety for an audit-grade product
• Build document-understanding pipelines for receipts, invoices, HUD statements, and forwarded emails — combining OCR, NLP, and structured extraction
• Train and serve transaction-classification models that map bank-feed and credit-card data to GL accounts, properties, and IRS categories
• Implement anomaly-detection systems that flag misclassified transactions, duplicate entries, and reconciliation discrepancies
• Partner with the full-stack team to integrate models end-to-end — model accuracy without good UX is wasted
• Set the standards for ML rigor in a financial product: evaluation, reproducibility, drift monitoring, and human-in-the-loop review

What we're looking for:

• 4+ years building production ML/AI systems — deployed, monitored, and continuously improved (research notebooks alone don't count)
• Strong Python proficiency; experience with PyTorch, TensorFlow, scikit-learn, or HuggingFace
• Hands-on experience with NLP, document understanding, or financial/structured ML
• Practical experience with LLM integration: prompt engineering, RAG, tool use, or agent frameworks (LangChain, LlamaIndex, or custom agent harnesses)
• Familiarity with OCR and document-processing tooling (Tesseract, AWS Textract, Google Document AI, etc.)
• Experience with cloud-based ML infrastructure (SageMaker, Vertex AI, or rolled-your-own on AWS/GCP)
• Strong intuition for evaluation: building eval sets, measuring model quality, and driving continuous improvement
• Comfortable owning the full ML lifecycle in a startup — data, models, deployment, and monitoring
• Bonus: experience building human-in-the-loop systems that learn from user corrections
• Bonus: experience operating LLM/agent systems in production with safety and grounding requirements
• Bonus: any exposure to financial or accounting data

Compensation: Competitive salary + equity. This is a foundational engineering hire — comp reflects that.

Location: Remote, with preference for candidates east of Denver (Central or Eastern time) to maximize collaboration with our US-based leadership and Moldova-based engineering team.

Our users are real estate investors, not accountants. You'll design interfaces that make complex financial data feel simple and actionable — dashboards, mobile receipt capture, tax summaries, and portfolio views. Proficiency in Figma and experience designing for fintech or SaaS preferred.

You'll own the full marketing funnel — content strategy, SEO, social media, paid acquisition, email campaigns, and partnerships with real estate communities. Experience marketing to real estate investors or in B2B SaaS growth is a strong plus.

We're a growing team and we're always looking for exceptional people. If you're passionate about real estate, fintech, or AI and think you'd be a great fit, tell us how you'd contribute.