Forward Deployed Engineer

3Pillar; Canada; On-site; CAD 140,000 - 200,000; Full time

Job summary

3Pillar seeks an experienced software engineer to own production AI system development for client engagements. You will design, build, test, and deploy AI-enabled data pipelines and integrations, focusing on reliability and scale.

You will work closely with client teams, applying retrieval architectures, agent orchestration, and LangChain/LangGraph to deliver production-grade solutions that drive adoption and measurable outcomes.

Qualifications

• 10+ years of professional software engineering experience.
• 2+ years building and shipping production AI systems.
• Depth in retrieval architecture (RAG) and agent orchestration patterns.
• Strong Python proficiency with production LangChain and/or LangGraph.
• Hands-on AWS across core compute, storage, and AI/ML services.

Responsibilities

• Own the last mile: deployment, monitoring, and user adoption.
• Build, test, and harden production-grade integrations against client systems.
• Run build-measure-learn cycles with client stakeholders to translate problems into scoped solutions.
• Act as a trusted technical counterpart to client engineering leads and executive sponsors; communicate tradeoffs in business terms.
• Contribute reusable patterns, connectors, and accelerators back to HelixAI for future engagements.

Skills

Python

LangChain

LangGraph

AWS

APIs

Field engineering

Tools

Bedrock

Lambda

S3

Job description

Embed on-site or in tight virtual proximity with client teams for the life of the engagement, and identify the highest-leverage workflow to build first, then the next.

Build, test, and harden production-grade integrations (APIs, MCP connectors, data pipelines) directly against client systems.

Own the last mile: deployment, monitoring, user adoption, and iteration based on real usage, not just proof-of-concept handoff.

Run build-measure-learn cycles with client stakeholders, translating ambiguous business problems into scoped technical solutions.

Act as a trusted technical counterpart to client engineering leads and executive sponsors; communicate tradeoffs in business terms.

Contribute reusable patterns, connectors, and accelerators back to HelixAI (Atlas, Nexus, AIRE) for future engagements.

What Good Judgment Looks Like Here

Simplifies complexity fast: turns a messy client environment into a clear, sequenced plan without waiting for a full spec.

Decides soundly under pressure and owns the call, rather than escalating everything or waiting to be told what to do.

Writes production code directly when that's what moves the engagement forward. This isn't an architecture-only or advisory seat.

Spots risk early and adapts, rather than discovering a blocker after it's already cost the engagement time.

Turns what's learned in the field into something reusable, not a one-off fix for one client.

Qualifications

10+ years of professional software engineering experience, including 2+ years building and shipping production AI systems (not prototypes).

Depth in retrieval architecture (RAG), agent orchestration/tool-use patterns, and evaluation pipeline design.

Strong Python proficiency, with production experience using LangChain and/or LangGraph.

Hands-on AWS experience across core compute, storage, and AI/ML services (e.g., Bedrock, Lambda, S3).

Demonstrated ability to work independently and ambiguously in a client-facing, field environment.

Hands-on experience integrating with enterprise systems via APIs, webhooks, or connector frameworks.

Prior experience in a forward-deployed, solutions engineering, or field engineering role.

Nice to Have

Experience with agentic SDLC methodologies and AI-assisted development tooling (Claude Code, Cursor, GitHub Copilot, or equivalents).

Familiarity with RBAC, audit logging, and governance requirements for regulated or enterprise clients.

Background in a specific vertical (financial services, asset management, healthcare, etc.).