Founding Engineer - Data

Mason; San Francisco (CA); On-site; USD 150,000 - 230,000; Full time

Job summary

Mason, a rapidly growing tech company in San Francisco, is building a company brain by turning decades of fragmented documents, databases, emails, and operational knowledge into connected data for AI agents.

You will design and build backend systems to ingest, organize, connect, and retrieve enterprise information, ensuring accuracy and context for AI tasks. This role blends backend engineering, data architecture, and experimentation.

Qualifications

• Four+ years of engineering ownership on production backends, data platforms, or complex data integrations.
• Strong programming and SQL skills with emphasis on system design and data modeling.
• Experience maintaining pipelines as sources, schemas, and volumes evolve.
• Comfort with messy, heterogeneous data and real-world problem solving.
• Ability to test approaches on real data and turn experiments into dependable systems.
• High ownership, curiosity about data work improving downstream product.

Responsibilities

• Turn fragmented company data into a trusted company brain for AI agents.
• Design and build backend systems to ingest, organize, connect, and retrieve knowledge.
• Own production pipelines and processing infrastructure with strong data modeling.
• Develop parsing, extraction, deduplication, and reconciliation for real-world data.
• Create storage, indexing, and retrieval systems preserving sources and access control.
• Implement observability to measure extraction quality, freshness, latency, and cost.
• Scale architectures to billions of rows and large document collections while enabling updates.

Skills

Backend engineering

Data pipelines

System design

SQL

Python

TypeScript

Data modeling

Knowledge graphs

Tools

PostgreSQL

Snowflake

Databricks

Amazon Redshift

dbt

LangGraph

Retrieval and memory

pgvector

Job description

Mason - Accelerating development of the built world

Mason is on a mission to accelerate the development of the built world. We are tackling some of the most pressing challenges of our time - housing shortages, energy constraints, and decaying infrastructure - all of which are exacerbated by the slow pace of physical development.

This is one of the largest, least-modernized markets in the world. Trillions in spending, and much of the work still runs on spreadsheets, documents, and manual processes.

We create AI systems that transform some of the largest firms in the world.

A core part of this work is building the company brain: turning decades of fragmented documents, databases, emails, and operational knowledge into connected data and context that AI agents can actually use.

We're growing fast - we have an 8-figure pipeline of projects and are looking for great engineers to join our team.

Our founding team comprises a repeat founder who exited to a Fortune 500, a former tech lead at Meta Superintelligence who co-created Meta AI, and a development director who built $2B+ of projects.

The Role

You'll be the person who turns a company's fragmented information into a company brain that AI agents can trust. You'll design and build the backend systems that ingest, organize, connect, and retrieve decades of enterprise knowledge.

The hard part goes well beyond moving data from one place to another. A property might have different names across systems. Documents can contradict one another. Important context can sit in a spreadsheet, a drawing, or an email thread. You'll figure out how to preserve those relationships and give an agent the right information for the task in front of it.

This role combines backend engineering, data architecture, and experimentation. You'll work closely with our CTO and the Infrastructure and Deployment engineers. You'll own production pipelines and processing infrastructure, with particular depth in data modeling, reconciliation, retrieval, and context quality.

What you'll build

• Ingestion and processing pipelines across databases, PDFs, policies, drawings, spreadsheets, and emails

• Backend services and shared data infrastructure that turn customer integrations into reusable capabilities

• Data models and entity relationships that connect information across fragmented systems

• Parsing, extraction, deduplication, and reconciliation methods that work on messy real-world data

• Storage, indexing, and retrieval systems that give agents relevant context while preserving sources and access permissions

• Evaluation and observability that make extraction quality, freshness, retrieval accuracy, latency, and cost measurable

• Processing architectures designed to grow to billions of rows and large document collections, with reliable updates and reprocessing

Tech stack and tools

Relevant experience includes:

• Languages and backend: Python, TypeScript, Node.js, SQL

• Data platforms and pipelines: PostgreSQL, Snowflake, Databricks, Amazon Redshift, dbt

• AI and agents: Anthropic and OpenAI APIs, Claude Agent SDK, Vercel AI SDK, LangGraph, MCP

• Retrieval and memory: pgvector, Cohere, Hindsight, RAG pipelines

• Workflows and compute: Inngest, Temporal, Blaxel, Modal

• Cloud and infrastructure: AWS, Docker, Kubernetes, Terraform

We value experience building production systems with these or comparable technologies.

What we're looking for

• At least four years of engineering experience, with meaningful ownership of production backend systems, data platforms, or complex data integrations

• Strong programming and SQL skills, with depth in system design, databases, and data modeling

• A track record building and operating pipelines that remain correct as sources, schemas, and volumes change

• Experience working with messy, heterogeneous data and resolving problems beyond the happy path

• Strong systems and algorithmic intuition, including tradeoffs around scale, correctness, incremental updates, and cost

• The ability to test an approach on real data, measure its limitations, and turn a promising experiment into a dependable system

• High ownership, high agency, and curiosity about how your data work improves the downstream product

• Experience with document processing, information retrieval, entity resolution, knowledge graphs, or LLM context systems is a strong plus

Team and culture

• CEO is a second‑time founder who sold his last company to Coinbase

• CTO was previously a tech lead at Meta Superintelligence and co‑created Meta AI

• COO has led over $2B in real estate development

• Backed by investors, operators, and major development firms

• Working directly with teams managing billions in AUM

• Small, intense, low‑ego team with substantial ownership from day one

• Regular board game nights, lots of laughs, no corporate nonsense