Applied AI Engineer

Levanta Labs is the data engineering, software development, and AI partner for high-growth B2B companies & SaaS startups. We design, build, and scale custom software platforms, internal systems, and AI solutions that drive measurable operational and financial impact.

You'll be the engineer who turns frontier AI into production systems deployed inside large organizations, embedded across departments, and woven into complex workflows. From rolling out enterprise AI platforms and the skills, prompts, workflows, and connectors that power them, to architecting bespoke AI applications end-to-end — choosing models, designing agents, picking frameworks, and shipping the orchestration that ties it all together — the engineering problems here excite us every single day.

Applied AI is how our clients compress weeks of work into hours, encode hard-won institutional knowledge and systems into agents their whole team can call on, scale expert-level judgment across departments, and empower their best people to focus on what only they can do. You'll work closely with client teams, founders, and our backend engineers from the earliest stages of an engagement, helping shape how AI lands in day-to-day operations and what gets built around it.

Roles & Responsibilities

* Take ownership of AI deployments end-to-end: scope → build → deploy → measure → iterate.

* Support rollout and configuration of enterprise AI platforms inside client organizations, from initial deployment through adoption.

* Partner directly with client teams to identify where AI creates leverage, then translate those needs into concrete skills, prompts, agents, and tooling.

* Design, build, version, and maintain libraries of reusable skills, workflows, and connectors that extend what AI agents can do.

* Architect bespoke AI applications and agents end-to-end: selecting models, defining tool access, structuring memory and context, and assembling the orchestration layer that ties it all together.

* Work across modern agent frameworks (Mastra, LangGraph, and similar) and build custom tools or MCP servers when off-the-shelf integrations don't go far enough.

* Apply engineering rigor to AI work (versioning, testing, evals, documentation, and clean rollback paths) and run structured evaluations to iterate on real usage data.

* Write and refine high-quality prompts and system instructions that consistently produce reliable, on-brand outputs.

* Collaborate with backend engineers on supporting infrastructure (retrieval pipelines, context management, vector search, knowledge graphs) when we need to go beyond what a platform delivers out of the box.

What You Offer (Requirements)

* First-hand experience building with LLMs (custom skills, agents, prompts, and tools) in any combination of jobs, side projects, hackathons, or open source. New grads welcome if you're deeply AI-native and have built things that go well past simple API calls.

* Strong proficiency in Python and comfort in TypeScript/JavaScript.

* Deep familiarity with the modern LLM ecosystem — frontier models, prompting techniques, tool use, agent patterns, and the tradeoffs between them.

* Experience with at least one modern agent framework (Mastra, LangGraph, or similar) and an understanding of when to reach for one vs. building from primitives.

* Demonstrated experience designing and shipping prompts, agents, or AI features that run in production (not prototypes).

* Strong software engineering fundamentals — version control, testing, code review, CI/CD — and a habit of applying them to AI work.

* Excellent written and verbal communication; comfortable working directly with client stakeholders to scope work and explain tradeoffs.

* A bias toward shipping, with the judgment to know when something is good enough vs. when it needs another pass.

* Comfort working in a fast-paced, client-driven environment.

Nice to Have

* Experience with the Model Context Protocol (MCP), tool use, or building custom integrations into AI systems.

* Background designing evals or building eval harnesses for LLM-based systems.

* Experience deploying enterprise AI tooling or copilots inside non-technical teams.

* Familiarity with retrieval pipelines, vector databases, or knowledge graphs.

* Experience routing across multiple model providers and reasoning about cost/latency/quality tradeoffs.

* Exposure to workflow automation or process mapping.

* Prior consulting, solutions engineering, or client-facing technical experience.

* CS degree or equivalent practical experience.

Tech Stack

Python, TypeScript/JavaScript, Node.js, Mastra, LangGraph, MCP, frontier LLM APIs (Anthropic, OpenAI, and others), PostgreSQL, vector databases, Git, AWS

What Success Looks Like

* Custom AI applications and agents you build are adopted, sticky, and materially change how teams work (not shelfware).

* The skills, workflows, agents, and connectors you ship are reliable, well-tested, properly versioned, and easy to extend.

* Prompts and systems you build produce consistent, high-quality outputs and degrade gracefully when they don't.

* You translate fuzzy problems into concrete AI capabilities quickly, without waiting for a perfect spec.

* You proactively measure what's working, surface what isn't, and iterate based on real usage data.

What We Offer

* Fully remote role (Canada, core hours 9am–6pm EST).

* Work at the frontier of applied AI on transformation projects spanning a wide range of industries.

* Direct, hands-on time with frontier AI systems and the people deploying them at scale.

* Mentorship from experienced engineers.

* The full weight of our organization behind your growth - paid certifications, conferences and events, intros and coffee chats with operators, founders, and specialists you want to learn from, and more.

* Clear growth path to senior roles with increasing ownership and responsibility.

* Performance bonuses tied to project outcomes.

* Gym membership stipend.

* Flexible paid time off.

* Direct collaboration with founders and a world-class technical team.

Work Environment

As a remote-first company, Levanta Labs is built for high performance and flexibility. We collaborate through Slack, Linear, Asana, Google Workspace, and Loom, maintaining a fast-paced but structured operating rhythm.

Equal Opportunity

Levanta Labs is an equal opportunity employer. We celebrate diversity and are committed to creating an inclusive environment for all team members. All qualified applicants will receive consideration without regard to race, color, religion, gender, gender identity, sexual orientation, national origin, disability, or veteran status.