AI & Emerging technologies

Senior AI Engineer (Full Stack)

Anyset Inc.

Montreal · Quebec · Canada

Share this job: X Facebook WhatsApp Save job : Senior AI Engineer (Full Stack)

What we're hiring

Senior AI Engineer (Full Stack) at Anyset. Montreal, QC (in office).

About Anyset

At the end of every construction project, general contractors have to organize thousands of documents from various systems, subcontractors, and the field so they can prepare that data for digital delivery to the owner. Today, that work happens almost entirely in email and spreadsheets. It's slow, and it usually end in a PDF binder.

Anyset was founded by construction engineers who lived with this problem on large infrastructure projects like roads, bridges, and manufacturing facilities. We're using AI to turn scattered data across project management systems like Autodesk and Procore into structured closeout, commissioning, and handover records that owners can actually use.

Role overview

We're building an agent harness for construction. A single project holds thousands of connected records: equipment, systems, locations, specifications, submittals, checklists, tests, issues, warranties, and the documents behind each of them. Our agents need to understand how these relate, find the right context in a very large project, and do real work on their own: read and fill documents, raise issues, chase people for missing information, and assemble the handover.

This is a hard systems problem, not a chatbot. You'll design how agents see a project, which tools they get, how context is selected and compressed, how work is split across subagents, and how runs survive failures and ask a person before they change something important. You'll work across the whole stack, from the database and permission model to the tools, the agent runtime, and the interface where people work with the agents.

What you'll do

  • Design and build the agent runtime: durable runs, tool calling, approvals, memory, and subagents that split long tasks into smaller ones.
  • Own context engineering: decide what an agent sees, retrieve the right records and documents from large projects, and keep context small and correct.
  • Model the relationships between construction entities and documents so agents can follow them, and expose them as well-designed tools.
  • Build and maintain our MCP server so customers can use Anyset from their own AI clients with the same permissions they have in the app.
  • Build evaluations that measure whether agents get real construction tasks right, and use them to compare models and prompts.
  • Ship full-stack features: TypeScript APIs, Postgres schema, background jobs, and React and React Native interfaces.
  • Work directly with customers and the founders to turn messy field workflows into product.

What we're looking for

  • 6 or more years of professional software engineering, with senior ownership of production systems.
  • Hands-on experience building LLM applications in production: tool calling, agents, retrieval, structured output, and evaluation.
  • A strong grasp of context engineering and its trade-offs: what goes in the window, what gets retrieved, what gets summarized, and how to tell when an agent is failing.
  • Strong TypeScript and Postgres skills, and comfort across backend, data modeling, and frontend.
  • Good judgment about system design: permissions, data integrity, failure modes, and cost.
  • Clear writing, and the ability to explain a technical decision to non-engineers.
  • Experience with multi-agent systems, MCP, document understanding, or construction and engineering data is a strong plus.

Our stack

TypeScript everywhere. Bun, Hono, and Mastra on the server, Postgres with Drizzle, Zero for sync, BullMQ on Redis, React and React Native on the client, and Gemini, Claude, and Grok models through Vertex AI, all running on Google Cloud.

Why Anyset

  • Hard, real problems: Agents that do multi-step work on large, connected construction data, not a demo.
  • Get in early: Work in person with the founding team in Montreal and shape the architecture from the start.
  • Competitive compensation: Competitive salary with eligibility for equity grants.
  • Make an impact: Help contractors finish jobs well and give owners a real, usable record of the buildings and infrastructure they operate.
#J-18808-Ljbffr

Reference: WJ-4483_1652569 · Posted 17 Nov 2026, 07:15 PM

Find more on GigBlow

This role is listed on GigBlow for discovery and search. Hiring decisions and applications are handled by the employer or their chosen application system.

AI Engineer jobs in other states

Browse jobs