Founding Engineer, AI Agents

Tyce; Toronto; On-site; CAD 120,000 - 180,000; Full time

Benefits offered by this job

Equity

Early-stage startup

Collaborative environment

Job summary

Tyce is hiring backend engineers to own the AI core of our product. You will ingest, understand, retrieve, and reason over customer documents, building AI-native context from complex PDFs and Excel files. We operate with FastAPI, PostgreSQL, Pydantic-AI, AWS, and a React/TypeScript frontend.

We seek mid-to-senior AI/engineering talent with strong ML fundamentals, production experience, and a go-getter mindset who can ship features daily and influence the AI roadmap.

Qualifications

• Mid-to-senior level AI/engineering experience.
• Strong applied ML fundamentals: model training/fine-tuning, embeddings.
• Background in search/ranking systems, NLP, or large-scale data/ML infrastructure.
• Hands-on experience building and evaluating LLM/RAG or agentic systems in production.

Responsibilities

• Own and evolve core parts of our AI platform, spanning search, retrieval, and multi-agent orchestration.
• Build a context/knowledge graph around our customers' knowledge bases: process very complex PDFs and Excel files into AI-native context representations.
• Improve our agent harness: agentic search and the unique tools that make up our platform.
• Build the evaluation harness that keeps quality high as we ship fast.
• Ship AI features day-to-day and help shape the AI roadmap.

Skills

Python

AI/ML

NLP

LLM/RAG

Search/IR

Distributed ML

Tools

FastAPI

PostgreSQL

Pydantic-AI

AWS

React Router

TypeScript

Job description

(\"Cracked engineers only\": Tyce is looking for backend engineers who ship fast and own outcomes.)

About Us

We\'re solving one of the largest problems in the physical world: public infrastructure that\'s too slow and too expensive to build. Construction is a $17 trillion global industry, and it runs on human expertise that doesn\'t scale, so we\'re building the AI that understands what to build and how to build it.

We start with the costliest problems general contractors face: the multi-million-dollar bidding and estimating decisions that off-the-shelf AI tools can\'t touch. Tyce is already live in production with industry landmark general contractors, estimating costs on multi-million-dollar builds and helping them win more bids.

We\'re building a product engineering team we believe can run a company worth a billion dollars. Competitive salary and equity.

Role Overview

You\'ll own the AI core of the product: how we ingest, understand, retrieve, and reason over customers' documents. This is where the hardest, most interesting problems live.

We\'re on a modern, lean stack: FastAPI, PostgreSQL, Pydantic-AI, Logfire, AWS, with a React Router / TypeScript frontend.

What You\'ll Do

• Own and evolve core parts of our AI platform, spanning search, retrieval, and multi-agent orchestration.
• Build a context/knowledge graph around our customers' knowledge bases: process very complex PDFs and Excel files into AI-native context representations.
• Improve our agent harness: agentic search and the unique tools that make up our platform.
• Build the evaluation harness that keeps quality high as we ship fast.
• Ship AI features day-to-day and help shape the AI roadmap.

What You Have

• Mid-to-senior level of AI/engineering experience.
• Strong applied ML fundamentals: model training/fine-tuning, evaluation methodology, embeddings, not just prompting LLMs.
• Background in one or more of: search/ranking systems, NLP, recommendation systems, or large-scale data/ML infrastructure.
• Hands-on experience building and evaluating LLM/RAG or agentic systems in production.
• Strong in Python; comfortable shipping and maintaining production systems.
• A go-getter mindset: personal agency, fast learning, strong verbal and written communication.

Nice to Have

Any of these is a strong plus, tell us your superpower:

• Built and evaluated an AI agent / RAG system for real-world performance.
• Retrieval/search background (embeddings, ranking, vector + keyword hybrid).
• Comfortable full-stack enough to ship your own features end-to-end.

What We Offer

• Competitive salary plus meaningful equity in an early-stage company.
• A founding-level seat on a 10-person product engineering team with direct influence over architecture and product direction.
• A close-knit and collaborative early-stage startup environment where every voice is heard and every opinion matters