Full-Time

Senior AI Engineer

Posted on 8/14/2025

Prophia

Prophia

11-50 employees

Automates lease data abstraction and management

Compensation Overview

$140k - $170k/yr

+ Equity

Company Does Not Provide H1B Sponsorship

Seattle, WA, USA + 3 more

More locations: Washington, DC, USA | Oakland, CA, USA | Chicago, IL, USA

Remote

Preferred locations are Washington D.C., Seattle, Bay Area, or Chicago.

Category
AI & Machine Learning (2)
,
Required Skills
LLM
Pinecone
Microsoft Azure
Python
AWS
Google Cloud Platform
Requirements
  • 5+ years total software, data science, and/or machine learning experience
  • Deep fluency in Python; strong systems and API design skills is a big plus.
  • Proven experience shipping LLM-powered features: prompt/program design, RAG pipelines, agents/tooling, evaluation and safety/guardrails.
  • Hands-on with vector databases and semantic retrieval (Pinecone/pgvector/weaviate) and embedding models.
  • Experience with cloud services (AWS/GCP/Azure).
  • Excellent communication; comfortable navigating ambiguity in a fast-moving startup.
Responsibilities
  • Design & implement GenAI systems: retrieval-augmented generation (RAG), prompt/agent orchestration, function/tool calling and grounding on Prophia’s CRE corpus. Choose patterns (prompting single LLM calls, context engineering, building agentic workflows) based on task, data, latency and cost.
  • Own LLMOps in production: build evaluation harnesses (quality, safety, regression), prompt/version management, telemetry, and model routing; monitor drift, latency, and cost.
  • Vector search & data plumbing: design chunking/embedding strategies; select and operate vector stores (e.g., Milvus, Weaviate, Qdrant, Pinecone, pgvector) with an eye on throughput, cost and hybrid search.
  • Serve at scale: implement high-throughput inference streaming UX, caching and concurrency controls; containerize and deploy to AWS with robust CI/CD.
  • Collaborate cross-functionally: partner with PM/design to turn requirements into technical plans; with engineers on integration with the main platform and with customer teams to instrument quality signals.
  • Hands-on prototyping: quickly validate ideas by using AI assisted coding; ship iterative improvements behind feature flags and experiments.
  • Technical leadership: mentor engineers, establish best practices for AI reliability and observability and help shape our AI architecture.
  • In the first 90 days, you will: Stand up an evaluation pipeline for our core GenAI tasks with baseline metrics and pass/fail gates. Ship one end-to-end LLM-backed workflow into production with telemetry and fallback behavior. Reduce latency and cost of a critical path by instrumenting caching, batching, or better retrieval.
Desired Qualifications
  • Experience with document AI (OCR, layout parsing), knowledge graphs or evaluation datasets specific to contracts/leases.
  • Experience with Docker and Kubernetes/AWS Elastic Container Service (ECS).
  • Prior exposure to privacy-preserving techniques and model red-teaming (e.g., prompt injection defenses).
  • Background in CRE or other data-rich, compliance-sensitive domains is a BIG plus.

Prophia offers an automated lease abstraction and management platform for commercial real estate. It converts static lease documents into a dynamic, living abstract that links to original sources so users can verify and analyze lease obligations. The platform aggregates many documents into a single source of truth and includes features like a self-updating stacking plan to keep data current, enabling enterprise-wide access and data-driven decisions. Prophia aims to reduce financial risk and manual errors in rent rolls and to help real estate professionals monetize and optimize lease data across the organization, supported by industry benchmarks and research.

Company Size

11-50

Company Stage

Series A

Total Funding

$15.3M

Headquarters

San Francisco, California

Founded

2018

Simplify Jobs

Simplify's Take

What believers are saying

  • Thomas Byrne joined board January 2026 to accelerate proptech scaling.
  • Direct Yardi and MRI integrations drive enterprise workflow adoption.
  • Retail AI features expand into underserved CRE vertical since 2023.

What critics are saying

  • CoStar erodes share by bundling abstraction into dominant platform.
  • Yardi launches native AI modules, capturing captive users in 6-12 months.
  • Generic AI like ChatGPT commoditizes abstraction in 18-36 months.

What makes Prophia unique

  • Prophia combines AI abstraction with expert validation for verified lease data.
  • Platform links abstracts directly to source documents for transparency.
  • Prophia Abstract delivers instant AI-powered lease summaries for brokers.

Help us improve and share your feedback! Did you find this helpful?

Benefits

Health Insurance

Dental Insurance

Vision Insurance

Unlimited Paid Time Off

Paid Holidays

401(k) Retirement Plan

Parental Leave

Bereavement Leave

Employee Referral Bonus

Growth & Insights and Company News

Headcount

6 month growth

0%

1 year growth

-4%

2 year growth

-4%
Prophia
Mar 23rd, 2026
How brokers are using AI to move faster - without risking accuracy.

How brokers are using AI to move faster - without risking accuracy. For years, brokers have lived in a constant tension: move fast, but don't miss anything. Clients expect quick answers - about lease terms, escalations, options, expirations - but the truth is those answers are buried in dense documents, scattered systems, and sometimes, outdated spreadsheets. So brokers do what they've always done: dig, double-check, and hope nothing slips through. Now AI has entered the picture. Tools like ChatGPT have made it easier to read and summarize documents. But there's a catch - one that matters a lot in commercial real estate. ChatGPT doesn't know your deals. It can interpret text, but it doesn't understand your portfolio. It isn't grounded in structured lease data. And it doesn't guarantee that what it tells you is complete, current, or accurate enough to stand behind in a client conversation. That's where a purpose-built solution changes the game. From "helpful" to actually reliable Prophia's AI Assistant, available through Prophia Abstract, was designed specifically for commercial real estate workflows. Instead of just reading documents on the fly, it operates on top of structured, audited lease data - so the answers you get aren't just fast, they're dependable. Ask a question like: * "When does this tenant's rent step up?" * "What options are coming up in the next 12 months?" * "Are there any unusual clauses in this lease?" And instead of digging through PDFs or second-guessing a generic AI response, you get a clear answer tied directly to your data. That's a big shift - from searching for information to simply asking for it. What this looks like in practice Inside Prophia Abstract, brokers are using the AI Assistant to speed up everyday work that used to take hours. 1. Instant lease answers during live conversations Whether you're on a call with a client or prepping for a meeting, you can quickly pull key details without breaking your flow. No more "I'll follow up after I check." 2. Faster deal evaluation Need to understand a new asset or tenant mix quickly? Instead of manually reviewing every lease, you can ask targeted questions and get a synthesized view in seconds. 3. Reduced risk of missed details One of the biggest risks in CRE isn't what you know - it's what you missed. Because Prophia's data is structured and validated, the AI Assistant helps surface critical terms that might otherwise be overlooked. 4. Less reliance on manual processes Spreadsheets, notes, and institutional knowledge only go so far. With AI layered on top of clean lease data, brokers can rely less on manual tracking and more on a system that's built to scale with them. Why this matters now The industry is at an inflection point. AI is everywhere - but not all AI is created equal. Horizontal tools will continue to improve. They'll get better at reading documents and answering general questions. But in high-stakes environments like commercial real estate, "pretty good" isn't good enough. Brokers need answers they can trust in front of clients, partners, and investors. That's why the combination of structured data + CRE-specific AI is so powerful. It closes the gap between speed and accuracy - something the industry has struggled with for decades. Built for CRE, recognized by REACH Prophia is a REACH Commercial graduate, part of the leading technology scale-up program backed by Second Century Ventures and the National Association of Realtors(R). That backing reflects a broader industry shift: purpose-built technology is what will define the next generation of CRE workflows. And AI is at the center of it. The bottom line AI isn't about replacing brokers - it's about removing the friction that slows them down. With Prophia's AI Assistant in Abstract, brokers can: * Get answers instantly * Trust the data behind them * Spend less time digging and more time advising Because in this business, speed matters - but accuracy is everything. And now, you don't have to choose between the two. Try it here free. Alejandra is a Customer Success Manager at Prophia, where she helps commercial real estate teams unlock the full value of their lease data. With over a decade of CRE experience - including roles at Prologis, Kidder Mathews, and Prometheus Real Estate Group - she brings firsthand insight into the operational challenges of...

Prophia
Jan 12th, 2026
Thomas Byrne Joins Prophia's Board of Directors

Thomas Byrne joins Prophia's Board of Directors. Prophia Appoints PropTech Industry Veteran Thomas Byrne to Board of Directors, Bolstering Strategic Vision and Planning for Next Phase of Growth San Francisco, CA - [January 12th, 2026] - Prophia, the leading AI-powered platform for commercial real estate data and document intelligence, today announced the appointment of Thomas Byrne to its Board of Directors. Byrne, a widely respected proptech executive and advisor, brings more than two decades of experience scaling category-defining platforms across commercial real estate. Byrne currently serves as Board Member and Executive Director at Dealpath, where he plays a role in shaping strategy and growth initiatives. Byrne also is an active investor and advisor to numerous leading and high growth proptech organizations. Most recently, Byrne held senior executive positions at VTS, where he served as General Manager of Marketplaces and VTS Market. In that role, he led the development and launch of the VTS Market platform and oversaw the suite of marketing solutions stemming from the VTS acquisition of PropertyCapsule, where Byrne had served as CEO. Earlier in his career, Byrne spent a decade at LoopNet where he served as President and Chief Operating Officer, helping to scale the company into one of the most successful venture backed proptech companies and profitable businesses on the Nasdaq. His leadership contributed to LoopNet's strategic acquisition by CoStar in 2012 for $900 million - one of the most notable transactions in proptech history. In addition to his operating experience, Byrne has held board and advisory positions with leading industry innovators including Homesnap, CrowdStreet, Rentlytics, and DotLoop, and holds an MBA from Harvard Business School and a BS from Georgia Tech. "Tom is one of the most experienced and trusted operators in proptech, and his perspective will be invaluable as Prophia accelerates its next phase of growth," said Cameron Steele, CEO of Prophia. "His track record of building - and scaling - category-defining platforms aligns perfectly with our mission to deliver the most reliable data foundation in commercial real estate." "Prophia has solved one of the hardest problems in commercial real estate: transforming decades of unstructured lease data into accurate, actionable intelligence," said Byrne. "I'm thrilled to join the Prophia board as the company continues to expand its footprint with exceptional CRE firms and unlock opportunities for owners, operators, and capital markets leaders." The appointment underscores Prophia's continued momentum as the company scales its platform, expands into new product categories, and strengthens its position as the trusted source of truth for CRE data. Cameron Steele is the visionary CEO and Co-Founder of Prophia with a 30-year career spanning global, competitive markets in Europe and the United States. After completing his undergrad at Stanford, he joined tech-stalwart Oracle, followed by a dozen years in investment/private equity industries, where he established...

Prophia
Dec 16th, 2024
The Ultimate Holiday Gift: Free Lease Abstraction, Instantly

As a gift to its industry, Prophia, Inc. is thrilled to announce a free, Alpha release of its Instant Lease Abstraction Generator-a groundbreaking tool designed to automate lease abstraction and provide immediate access to critical lease data; at no cost.

Prophia
Oct 16th, 2023
Announcing: The New Deals Report By Prophia

That's why Prophia is excited to introduce its revolutionary "Deals Report," an innovative tool that is set to redefine how CRE professionals manage and leverage leasing data.

Prophia
Aug 24th, 2023
What's the Real Cost of Missing a Critical Date in Your Lease Data?

Despite these unexpected conditions, forward-looking teams like Lionstone Investments, Nuveen, RXR, and MetroNational have partnered with Prophia to enhance data management, prevent portfolio management errors, and boost ROI.

INACTIVE