ACL Digital
Website:
acldigital.com
Job details:
JOB DESCRIPTION
AI Engineer (GenAI / Agentic Systems)
Role Overview
This role owns the design and implementation of production-grade AI capabilities across four tightly integrated streams: (1) AI agents and orchestration, (2) context and retrieval (RAG), (3) prompt and policy management, and (4) evaluation and continuous improvement.
The AI Engineer works across these domains to ensure coherence, reliability, and governance of AI behavior in an enterprise environment. While sprint focus may vary by stream, accountability spans the full AI lifecycle.
Core Responsibilities
1. AI Agents & Orchestration
- Design and implement AI agents, tools, and orchestration workflows for enterprise use cases, including integration with Inception-built agents and orchestrators.
- Define and enforce agent action boundaries including allow/deny lists, least-privilege execution, and safe tool usage.
- Implement human-in-the-loop (HITL) controls for high-risk operations, data modifications, and external system interactions.
- Design multi-agent coordination patterns including delegation, handoffs, and conflict-free execution.
2. Context, Retrieval & Knowledge (RAG)
- Design and optimize retrieval pipelines including chunking, indexing, hybrid search, and query reformulation.
- Ensure permission-aware and lineage-aware retrieval aligned with enterprise data access controls.
- Improve grounding quality by reducing hallucinations and unsupported outputs.
- Support knowledge lifecycle including updates, refresh strategies, and staleness handling.
3. Prompt, Policy & Workflow Management
- Develop and maintain version-controlled prompts, policies, and workflow templates across environments.
- Implement change management including approvals, rollback mechanisms, and release alignment.
- Translate business and safety requirements into enforceable prompt and workflow constraints.
4. Evaluation, Monitoring & Continuous Improvement
- Design and maintain evaluation datasets, scoring frameworks, and automated evaluation pipelines.
- Track performance metrics including accuracy, drift, tool usage, and grounding quality.
- Integrate observability tools (e.g., Azure Monitor, Langfuse) for tracing and diagnostics.
- Continuously improve AI behavior through structured feedback loops and documented changes.
Required Qualifications
- 6+ years of experience in software or ML engineering.
- 4+ years hands-on experience building production-grade LLM, agent, or RAG systems.
- Strong Python skills and experience with APIs, data pipelines, and system integration.
- Strong testing mindset for non-deterministic systems.
- Demonstrated ability to work autonomously while maintaining a team-oriented and results-engaged mindset.
Preferred Qualifications
- Experience with Azure OpenAI and Microsoft-aligned AI tooling.
- Familiarity with agent protocols (e.g., MCP, A2A) and enterprise data governance (e.g., Purview).
- Experience building evaluation frameworks for hallucination, tool accuracy, and multi-step workflows.
- Experience working in enterprise or regulated environments.
Click on Apply to know more.