Founding Head of Engineering - cmdr

Help build the AI-native loan servicer.
Remote, U.S.

About cmdr

Servicing is the bedrock of lending. Every loan is serviced for its entire life. Someone collects the payments, tests the covenants, manages escrow, tracks insurance, spreads the borrower's financials and reports to investors. In the U.S., $5.1 trillion of commercial real estate debt needs servicing, and $1.5 trillion of it comes due in 2026 and 2027. Most of that work runs on platforms built in the 1980s and 1990s. Those platforms record what happened and leave everything else to people working from spreadsheets and inboxes. Covenant terms sit in PDFs that nobody re-reads until something goes wrong.

cmdr is building an AI-native loan servicer. We acquire servicers whole: their books, their people and their client relationships. Then we run every loan they hold on one live, sourced record. Our first acquisition, a commercial loan servicer with roughly $10 billion of loans under management, is in advanced discussions.

The record does the work that servicing depends on. We extract every term and covenant from the loan agreement, and every value carries the words it came from. AI reads, code computes, and a person confirms every exception. Each covenant becomes a tracked obligation: what the borrower owes, how often, what proves it, and the threshold. Obligations are checked as they happen, and a person is accountable for every decision. Agreement reading and human review run in production today.
The prize is larger than servicing. Servicing creates the live, sourced record that every other lending workflow runs on. We start with commercial real estate. Next comes underwriting and portfolio management on the same record, then private credit, auto, personal and student loans, and eventually finance beyond lending. We think autonomous finance starts here.

Why cmdr

• A growth model that compounds. We buy the servicer, not its clients. Every acquired book moves onto one platform, so improvements ship once, and what one book teaches, the next inherits.
• A working product on real loan files. Agreement reading and human review run in production, and we are building with lenders under NDA. You start with a working system and real books to run it on.
• A founder who has taken real estate public. Drew Sterrett, our founder and CEO, co-founded and led LEX, which took individual commercial buildings public through single-asset IPOs.
• Studio backing. Stackpoint, the Boston venture studio started by the co-founders of Notarize (now Proof) and Convene, committed our founding capital and is our operating partner. Its co-founders have created more than $2 billion in combined equity value. An institutional co-investment round is in progress.
• Real influence from day one. You will shape strategy, product, culture and execution from the start.

The Role

You'll be cmdr’s technical leader. You'll lead engineers at mixed levels, from early-career to principal, and shape how the team grows from there.

This is a hands-on, player-coach seat that spans architecture, code, evals, and how the team operates. You'll take the first version of the platform, built with Stackpoint's studio team. Together with the founding engineers, you'll set the technical foundation every future engineer builds on.

What You'll Drive

1. Own the architecture and the technical bar

• Lead system architecture, data models, and platform direction with the engineering team, and make the call when the team doesn't converge.
• Decide what to build, buy, or integrate, and when a prototype has earned the investment to scale.
• Keep the platform reliable and cost-efficient as volume grows and we expand to new agencies and states.
• Set the engineering standards the team works to: code review, testing, and deployment.

2. Ship agentic AI that holds up on real transactions

• Lead how the team builds extraction, classification, and retrieval over messy loan and lease documents, including integrations with the legacy software, portals, and email- and file-based workflows the industry still runs on.
• Define the evaluation system and guardrails: measure accuracy and confidence before anything ships, and decide what work goes to human review.

3. Build the human-in-the-loop product

• Lead the build of the internal operations console where loan servicers review, correct, and approve AI output all day, with their corrections feeding back into evaluation.
• Shape the client-facing experience servicers use to board loans, monitor obligations, and act on results.
• Make PII and sensitive transaction data a design constraint across both, with access control, audit trails, and tenant isolation.

4. Turn discovery into what gets built

• Partner with the Head of Product & Operations to translate operator and customer discovery into scoped, sequenced engineering work.
• Join operator and customer discovery so engineering decisions start from how the work gets done today.

5. Build and lead the engineering team

• Hire and grow the engineering team, and shape how it's organized and how it works as cmdr scales.
• Mentor early-career engineers and give senior engineers real ownership.
• Lead how the team uses AI to plan, build, review, and ship, and keep refining it as the tools change.

What Success Looks Like

• Onboarding the next agency or workflow is configuration, not a project.
• Automation and cost per file improve release over release, and the team can prove it in production data.
• Operators trust the review console, and correction time keeps falling.
• The team ships on a steady cadence without you as the bottleneck.

What You Bring

An engineering leader who still builds. You've led and grown engineers at different levels while staying hands-on in architecture and code, including meaningful production code in the last two years. AI coding agents and tools are core to how you work.

Production agentic AI. You've shipped LLM-powered workflows that real users depended on, and you know how to make them reliable: evals, structured outputs, retrieval, prompt design, and model and cost tradeoffs.

Full-stack and data range. You work across a Python or Node backend, a TypeScript/React frontend, relational and vector databases, and AWS or GCP. You've built pipelines that turn unstructured documents into structured data, integrated with legacy systems that lack clean APIs, and handled PII in a regulated setting.

Product sense for operational users. You've built internal or operator-facing tools for high-volume work, and improved them based on real usage.

A clear communicator across technical and non-technical teams. You've worked alongside product or operations leaders to turn discovery into plans engineers could execute, and explained technical risk and tradeoffs to non-technical leadership in plain terms.

Startup DNA. You've worked in ambiguous, resource-constrained environments, shipped quickly without lowering the quality bar, and moved between strategy and execution as the week required.

Experience

• 10+ years of software engineering experience, particularly in product-focused, full-stack roles.
• Demonstrated success as an engineering leader in a fast-paced startup or high-growth environment.
• At least one product taken from a blank page to production and scaled past its first customers.

This role is designed for a builder. If your recent experience is mostly managing managers, leading large-org transformation, or advising rather than shipping, it's unlikely to be the fit.

What Makes This Role Unique

• Ground truth from day one. Live orders, real volume, and operator corrections that become your evaluation set. Most AI startups spend their first year trying to get data like this.
• Founder-level scope. Real ownership and influence over the product, the platform, and the team you build.

An Ecosystem of Exceptional Builders

Joining cmdr also means joining Stackpoint's tight-knit ecosystem of founders, product leaders, and operators building alongside you. Engineering leaders across the portfolio are peers, not silos:

• Regular cross-portfolio sessions with founders and product and engineering leaders
• Shared learning on AI, GTM, hiring, and scaling challenges
• Direct access to other builders solving similar 0→1 problems in parallel, and pattern recognition across multiple early-stage companies

We value judgment, speed, and substance over process theater.

Compensation

• Competitive salary calibrated for founding team members
• Significant equity with asymmetric upside if we execute
• Comprehensive health benefits

If you read this and want to build the platform and the team behind the AI operating layer loan servicing operations run on, we'd like to talk.