Senior Software Engineer, Agentic AI Products

What we're looking for

We are building a suite of AI-native tools that will change how Leo Berwick delivers advisory work and how our clients engage with us.

We are hiring a Toronto-based Senior Software Engineer to join our founding in-house engineering team and build products from the ground up. This is a hands-on, individual contributor role for someone who writes code, solves hard problems, and gets things done.

You won't just be building wrappers around LLMs. You'll architect the 'harness' - the orchestration middleware that manages agentic workflows, handles complex context packing, and integrates frontier models with enterprise data to automate high-stakes M&A workflows.

You get the pace and ownership of an early-stage product team, backed by the domain expertise, client relationships, and built-in distribution of an established firm, with direct access to firm leadership. Our product is already proven with internal users, so you spend your time building and shipping, not searching for product-market fit.

This is a hybrid role, with a minimum of two days per week in our Toronto office location.

You’re the right fit for this role if you –

• Want to build a new product from the ground up and own what ships
• Thrive on ambiguity and bring a bias toward action: you figure out how to get things done
• Want to shape the architecture, technology choices, and engineering culture from day one
• Enjoy working directly with non-technical subject matter experts and turning their knowledge into product
• Want your work in the hands of real users quickly, with the potential to see internal tools evolve into client-facing, commercial SaaS products

What you'll do

Design and build agentic products

• Design and build agentic features and workflows, such as document and data extraction, multi-step engagement workflows, and governance controls, from prototype to production, moving toward 'automated associates' that handle complex tasks like document triage, automated agentic pipelines, and project management.
• Integrate LLMs, tool-calling, retrieval, and orchestration patterns into reliable, secure, and observable systems.
• Design and implement robust data/domain models that support automated reporting and memo generation via Model Context Protocol (MCP) and other structured data exposure layers.
• Write high-quality, production-grade code across the stack.

Shape the platform

• Help define the architecture and technology stack for a new, modern platform, and build it from the ground up.
• Contribute to engineering standards, code reviews, testing, scalability, manageability, and security and governance practices for AI-enabled systems.

Collaborate as part of a growing team

• Work alongside senior technology advisors and subject matter expert builders embedded full-time in the product, iterating quickly on feedback from internal users.
• Work in small, focused pods with in-house and external engineers, sharing knowledge so team capability compounds over time.
• Take part in technical evaluations, including case-study style discussions, and help hire other engineers as the team grows.

What you bring

You are a senior engineer with 7–15 years of professional software engineering experience, a strong academic foundation in computer science, engineering, or mathematics, and a proven record of driving technical delivery in complex, high-stakes environments.

• A track record of shipping. You've built enterprise SaaS or B2B products, and in the last 24 months you shipped an Agentic or LLM-powered product you contributed to hands-on. You can point to the impact, whether GitHub repositories or documented SDLC improvements. Startup or founding-team experience is an asset.
• Agentic systems depth. You've built or worked within an AI "harness", the middleware that orchestrates agents (Fireworks AI or similar orchestration layers). You understand the full LLM lifecycle, well beyond basic prompting.
• Sound architectural judgment. You can articulate the trade-offs between domain-driven design and monolithic architectures, and you know orchestration patterns such as enterprise service buses. You favour a "buy-first" strategy and put engineering effort into proprietary IP rather than solved problems like document extraction.
• Full-stack proficiency. You are strong in React, TypeScript, and Java, and comfortable picking up other languages, such as Python, as the work demands.
• Business-to-workflow translation. You can turn complex business processes, such as document triage, redlining, and spreadsheet organization, into automated agentic workflows.

Compensation

This is a permanent full-time role with a base salary range of $160,000 –$180,000 CAD. Placement within the range will depend on relevant experience and scope of responsibility. 
Vacancy Statement

This posting reflects a new vacancy.

Use of Artificial Intelligence in Hiring

Leo Berwick uses artificial intelligence tools to screen applicants for this role.

Equal Opportunity Statement

Leo Berwick is an equal opportunity employer committed to building a diverse, inclusive and accessible workplace. We welcome applications from all qualified individuals and do not discriminate on any ground protected under applicable human rights legislation. Accommodations are available upon request at any stage of the hiring process, in accordance with applicable accessibility and human rights legislation.

Please contact careers@leoberwick.com to request an accommodation. Information received relating to accommodation will be kept confidential and used only to support your participation in the hiring process.