Job Specifications
About Muro AI
Muro AI is transforming how the $2T construction industry plans and builds. Founded by Cornell alumni, ex-founders, and former McKinsey operators, we’re building AI agents that automate the most complex, manual, and costly phase of construction: preconstruction.
We move fast, build with conviction, and obsess over delivering real impact to the people who build our world. If you want to shape how the next century of construction gets built, this is where it starts.
Who You Are
A true builder with 4+ years of technical experience. You’ve lived the 0→1 chaos - maybe you co-founded something, or led a team through the trenches. You know that the good (fast growth) and the ugly (2 AM patches) come with the territory. You’re scrappy, curious, and energized by uncharted problems.
Why It Matters
Shape a category. No one owns “pre-con AI” yet — we’re charting that map.
Bridge two worlds. Effect changes in an estimator’s spreadsheet with multimodal models that understand construction documents.
Drive impact. Your AI agents will help contractors close multi-million-dollar bids and save hundreds of hours per project.
Your Mission
Build agentic workflows that work. Create agents to complete multi-step construction tasks. Implement tool use for Procore APIs, Excel manipulation, and document generation. Build agents that read plans, update budgets, and generate RFIs autonomously.
Ship production RAG systems. Process 1,500-page plan sets. Build pipelines that understand construction specs and drawings. Handle messy, real-world documents - scanned, rotated, hand-marked. Deliver semantic search that actually finds the door schedule on page 347.
Embed with customers. Sit with estimators during bid week. Understand workflows, then turn “we manually compare 50 subcontractor bids” into an intelligent agent.
Own the AI stack. Python services orchestrating model calls. Celery workers for long-running jobs. Vector DBs for semantic search. Context engineering, evaluation frameworks, and monitoring for quality. Dockerized workloads on GCP infra.
Make AI reliable. Build evaluation datasets from real construction docs. Collaborate with SMEs to catch hallucinations and make models accurate enough for million-dollar decisions.
Tech Stack
AI/ML: LLM APIs, Agentic frameworks, RAG pipelines, Vector DBs, Langfuse
Backend: Python (FastAPI), Celery, Node.js/TypeScript
Infra: GCP, Docker, Supabase
Data: PDF parsing, spreadsheets, computer vision, unstructured data pipelines
Who Should Apply
We want hands-on builders with 4+ years of experience delivering applied AI to real users. You might be:
A Full-Stack Engineer hacking with LLMs who wants to go deeper
A Forward Deployed Engineer who wants to build AI, not just integrate it
An ML Engineer who wants to ship products, not just models
A Data Scientist tired of notebooks who wants to ship production code
A Founding Engineer or ex-CTO ready for the next challenge
You'll Stand Out If You Have
Built and deployed an AI feature with real users
AEC industry experience
Worked with computer vision or advanced OCR pipelines
Fine-tuned a model with a custom-built datasets
Roles involving coding + customer interaction (FDE, Solutions Engineer, etc.)
Prior founding-team or 0→1 experience
Real Talk
You’ll own customer and agent implementations end-to-end.
You’ll ship to production within days, not weeks.
Specs will be half-baked - you’ll fill in the blanks.
You’ll wear many hats: dev, architect, product, recruiter, and support hero.
If you love making AI solve real problems for real people, you’ll thrive.
Location: In-person in San Francisco
Compensation: $120K – $250K base + competitive equity