Lead Software Engineer

Lead Engineer — Vensuris Labs Full-time · Remote (Eastern Standard Time prefered) About Vensuris Labs Vensuris Labs is the software arm of The Vensuris Group. We build agentic AI products for real, operating businesses — starting with ShopPros (an agentic operating layer for independent auto-repair shops) and expanding into a small portfolio of vertical SaaS bets, each anchored by proprietary data and deep subject-matter expertise. We run lean and move fast on a modern agentic stack (Claude Code, Next.js, Python, Postgres, Railway, Vercel) — a small, high-leverage team where one engineer's work compounds across the whole portfolio. This is a chance to come in early, set the technical bar, and grow the engineering org from the ground up. The role We've already founded the core platform, the verticals, and the SME relationships that connect us to live data and industry intel. Your job is to build it out — from early stage to MVP and into production. You'll take ownership of hardening, LLM-cost optimization, and build capacity across the platform, while working alongside Ricky (Head of Product & Technology), who stays on product, architecture, and technical leadership. Concretely, in your first 90 days you will: • Pair on the agentic boilerplate's patterns until you can wield and extend them • Own the hardening + LLM-cost backlog (model tiering, caching, context management) • Set the code-quality bar — testing, observability, security — that the team scales on • Lean into ShopPros (our largest base and most operational pressure), while Ricky validates the next verticals What you'll do • Build agentic AI products end-to-end: LLM orchestration, tool-calling, multi-agent loops, RAG, and evals • Work across a modern full stack — TypeScript / Next.js on the front, Python on the backend, Postgres for data, Railway + Vercel for infra • Own quality: automated tests, observability (Langfuse), security, and strict multi- tenant data isolation• Keep LLM cost under control — the single largest variable cost line in an agentic product • Turn one-off solutions into reusable patterns in the shared boilerplate that compounds across every product Must-haves (the gate) 1. You've shipped agentic AI in production — real systems with LLM orchestration, tool-calling, and evals. Not tutorials, not prompt-only demos. Non-negotiable. 2. Strong backend / full-stack engineering — fluent across the stack above (or clearly adjacent and demonstrably fast to ramp). 3. You own quality — testing, observability, security, and tenant isolation are instincts, not afterthoughts. 4. LLM cost instinct — you've tuned model tiering, caching, and context to keep spend sane. Who you are Hungry. Smart, with high learning velocity. Humble and coachable. Trustworthy. Highly motivated, with a long-horizon commitment to the mission. You want to build — and you measure yourself by what ships and works, not by title or tenure. Compensation • Market-salary compensation, full-time, plus benefits. • Early-stage compensation incentives. As an early hire, you'll share in the upside - there will be early-stage compensation incentives for the team that builds this out.

Build and scale agentic AI products from MVP to production, focusing on LLM orchestration and multi-agent loops. Take ownership of platform hardening, cost optimization, and establishing code quality standards for the engineering organization.

Must have a proven track record of shipping agentic AI systems in production with real-world LLM orchestration and evals. Requires strong full-stack engineering proficiency in Python and TypeScript, with a deep instinct for security, observability, and cost management.

• Market-salary compensation • Early-stage compensation incentives

• Next.jsSouhaitée
• Tests automatisésSouhaitée
• PostgreSQLSouhaitée
• TypeScriptSouhaitée
• PythonSouhaitée

Relevées dans la description du poste. Confirmez les exigences importantes ci-dessus.

• LLM Orchestration
• Tool-calling
• Multi-agent loops
• RAG
• TypeScript
• Next.js
• Python
• Postgres
• Railway
• Vercel
• Langfuse
• Observability
• Multi-tenant data isolation
• LLM Cost Optimization
• Automated Testing
• Security

• Software
• Technology
• Engineering
• Management & Leadership
• Data & Analytics

Expérience minimale

5+ ans

Postuler avant le

5 sept. 2026

Langue de l’offre

anglais

Heures de travail

40 heures par semaine

Exigences de lieu

Country, Greater Toronto Area, Canada

Niveau d’expérience

Mid-Senior level

Mode de candidature

La candidature directe est offerte