Principal AI Platform Engineer in Toronto

At Epiq, your work contributes to complex, global legal outcomes. You’ll join a values-driven community where integrity guides decisions, relentless service sets the bar, and we thrive on big challenges together. We invest in your growth with enterprise-wide learning and mobility. We celebrate who you are, and we respect life beyond work with flexibility that’s recognized externally. Enabled by modern platforms and AI, you’ll do the most meaningful work of your career and see your impact at scale.

Epiq supports some of the world's largest and most complex legal matters, where document collections range from thousands to millions of records. As a Principal AI Platform Engineer, you will design and build the platform our AI agents and accelerators run on, including orchestration, retrieval, evaluation, observability, and the inference path that keeps all of it fast and affordable at production scale. This is a senior individual contributor role. You will write production code; own significant components end to end. You will set the technical direction that other engineers build on. You will also mentor engineers across our Toronto, US, and India teams through design reviews, code reviews, and pairing.

You will work across two product areas:

• AI Agents: Conversational systems that enable legal professionals to ask complex questions and receive evidence-based answers grounded in case data.
• AI Accelerators: AI-powered workflows that automate large-scale document review tasks including summarization, translation, transcription, language detection, and OCR.

Together, these solutions help legal teams reach defensible outcomes faster while reducing review costs and improving quality.

What You'll Do

• Platform Engineering: Build and own the platform capabilities that power AI agents, including orchestration, retrieval, model routing, tool integration, and reliability. Design retrieval systems across large-scale legal document collections. Develop shared platform services including prompt management, caching, rate limiting, and cost attribution. Define stable APIs and data contracts for product teams.
• Evaluation & Quality: Develop evaluation frameworks that measure accuracy, grounding, citation quality, and overall system performance. Build observability, tracing, and replay capabilities that enable rapid troubleshooting and continuous improvement. Establish quality gates and monitoring to identify regressions before they reach production.
• Performance, Scale & Cost: Optimize latency, throughput, and scalability across both interactive and batch workloads. Drive decisions around model selection, inference strategies, caching, and routing while managing cost as a first-class engineering metric.
• Reliability & Governance: Build highly reliable systems through monitoring, alerting, incident response, and operational excellence practices. Implement governance, auditability, and human-in-the-loop controls required for legal and regulated workflows. Ensure compliance with security, privacy, and data residency requirements.
• Technical Leadership Without Direct Reports: Write the design docs and ADRs for consequential platform decisions, including the trade-offs you rejected. Raise the engineering bar through code review, pairing, and mentoring. Grow other engineers’ judgment, not just their output. Partner with Product Management, legal domain experts, and Review Operations to turn ambiguous problems into shippable scope.

Required Qualifications

• Bachelor's or master’s degree in computer science or a related technical field, or equivalent practical experience.
• 7+ years building production software, with a track record of owning systems end to end.
• Experience building production AI/LLM systems, including retrieval-augmented generation and agentic workflows.
• Strong Python. Comfortable in a TypeScript codebase.
• Experience with distributed systems, asynchronous and event-driven architectures, and large batch data processing.
• Demonstrated ownership of production reliability: instrumentation, on-call, incident response, and performance work driven by data rather than intuition.
• Clear written communication. You can explain a design decision and its trade-offs to engineers and to non-engineers.

Preferred Qualifications

• Orchestration frameworks such as LangGraph, LangChain, MCP, etc.
• LLM observability and evaluation tooling such as Langfuse, OpenTelemetry, etc tracing.
• Vector databases, search infrastructure, and information retrieval fundamentals.
• Hallucination detection, grounding verification, and benchmarking methodology.
• Model fine-tuning, quantization, or inference optimization.
• React and modern frontend work for internal evaluation and debugging tools.
• Legal technology, eDiscovery, compliance, investigations, or document review.
• Multi-tenant SaaS platforms with strict data isolation requirements.
• Experience with PostgreSQL.

Epiq is committed to equal employment opportunity laws and prohibits unlawful discrimination based on any impermissible bases. If you need assistance or an accommodation during the application process because of a disability, it is available upon request.