Lead AI Integration Engineer

Lead, AI/ML Engineering

As a Lead, AI/ML Engineer at 3

Pillar, you will lead end-to-end AI solutions for our clients. You will work on complex AI/ML problems, contribute to architecture decisions, and ship agentic, LLM-based systems and ML solutions in production.

This role sits at the intersection of backend engineering and applied AI. You are someone who can seamlessly connect APIs, orchestrate data flows, and build Retrieval-Augmented Generation (RAG) systems, all while maintaining rigorous data privacy standards. In an era defined by AI, you will be the force that turns theoretical AI concepts into production-ready features.

Key Responsibilities

• Design, build, and deploy intelligent backend services using Python and Azure AI Foundry.
• Consume Force Account outputs by API, ensuring the system can accurately display and act on data classification and extraction.
• Engineer and optimize Retrieval-Augmented Generation (RAG) pipelines over Standard Operating Procedures (SOPs) to provide contextual, accurate, and actionable insights to users.
• Deliver assistive intelligence directly on the case, seamlessly integrating AI outputs into the end-user's workflow.
• Design and strictly enforce a redact-before-model discipline, ensuring that all Personally Identifiable Information (PII) is stripped from payloads before data ever reaches a foundational model.
• Collaborate with product, design, and engineering teams to shape AI features and translate business goals into technical execution.
• Proactively identify AI-driven automation and product improvement opportunities, building prototypes to prove value.
• Mentor junior engineers and champion secure, responsible AI coding practices within your team.
• Troubleshoot AI-specific technical challenges, including prompt injection risks, model latency, and data leakage.
• Guide team members on model selection, evaluation, fine-tuning, and Gen AI usage.
• Actively evaluate emerging LLM tooling, agentic frameworks, and open-source developments for applicability to client engagements.
• Own the architecture of the team's agentic and LLM systems — agent orchestration, tool integration (e.g., Model Context Protocol), retrieval, and human-in-the-loop design.
• Design and implement evaluation frameworks for LLM and agentic systems — covering correctness, boundary, and intent checks — as a standard part of delivery.
• Drive adoption of the agentic SDLC across the team: select and roll out AI engineering tools, define usage and governance standards, and measure their impact on delivery velocity and quality.
• Establish responsible-AI and evaluation frameworks for agentic systems — guardrails, safety testing, human oversight, and model governance.

Minimum Qualifications

• A minimum of 8+ years of experience in software engineering, with at least 2+ years of hands-on experience integrating LLMs and building AI-driven applications.
• Demonstrated experience architecting and deploying LLM-based or agentic systems in production, including prompt design, retrieval architecture, and tool/function calling (e.g., MCP)
• Hands-on experience with leading LLMs and APIs, including OpenAI, Google Gemini, and Anthropic Claude.
• Experience evaluating and selecting open-source and proprietary LLMs (e.g., OpenAI via Azure AI Foundry) based on accuracy, latency, scalability, cost, and overall performance for production AI solutions.
• Knowledge of vector stores, embeddings (Lang Chain, Llama Index, etc.).
• High level of English proficiency required to interact with a globally-based development team and client stakeholders.
• Deep expertise in Python for backend services and AI integration (replacing traditional Java/C# requirements).
• Demonstrated, hands-on experience utilizing Azure AI Foundry to build, deploy, and manage AI models and agents.
• Proven track record of building and managing complex API integrations for data ingestion, extraction, and classification.
• Strong understanding of security engineering, specifically implementing PII redaction, anonymization pipelines, and secure data handling in AI workflows.
• Experience delivering well-tested, scalable, secure, and performant enterprise-level systems that achieve client business outcomes.
• Stron…