Senior AI Engineer - Agentic Systems
Senior AI Engineer at Gloat (Tel Aviv) designing, building, and shipping production-grade LLM agents for its workforce orchestration platform. Day to day: multi-agent orchestration, RAG over a knowledge graph, tool/function calling (MCP), evaluation harnesses and guardrails, and deploying model-agnostic agents into Teams, Slack, and Copilot using Python and frameworks like LangGraph.
- Compensation
- Not disclosed
- Location
- Tel Aviv-Yafo, Tel Aviv District
- Work setup
- Hybrid
- Seniority
- Senior
- Employment
- Not specified
- Experience
- 5+ years
- Category
- Ai engineering
- Posted
Skills
Must have
- 5+ years building production software
- Proven experience building and shipping LLM agents to production — not just demos or prototypes
- Hands-on with at least one agent orchestration framework (e.g. LangGraph, LangChain, AutoGen, CrewAI, Semantic Kernel)
- Strong Python and solid software engineering fundamentals
- Prompt engineering paired with systematic, measurable evaluation of LLM outputs
- Experience with tool use / function calling and integrating LLMs with external systems
- Track record deploying, monitoring, and maintaining AI in production (cloud, CI/CD, observability)
Nice to have
- 2+ years hands-on with LLMs / generative AI
- Practical experience with RAG, embeddings, and vector databases (e.g. pgvector, Pinecone)
- Experience with MCP, agent memory, and planning/reasoning patterns
- Background in HR tech, people data, or skills ontologies
- Knowledge graph / graph ML experience (knowledge graphs, GNNs)
- Responsible AI: bias evaluation, guardrails, and AI governance
- Experience working across multiple model providers (e.g. Anthropic, Google, IBM watsonx)
Full description
Gloat is the AI-native system of action for workforce orchestration. Our agentic HR platform lets enterprises build intelligent workforce agents that understand their people, respect their policies, and work where teams already are — Microsoft Teams, Slack, Google Chat, and Copilot, with no new portal or change management.
The platform is powered by Loomra, our semantic layer that connects people, jobs, and skills through a knowledge graph, deep HCM integration (Workday, SuccessFactors, Oracle HCM), a governance engine, and semantic retrieval — with a full audit trail behind every agent decision.
Trusted by 100+ enterprises across 112 countries, empowering 5M+ employees. This role sits at the center of that effort.
We're looking for a Senior AI Engineer to design, build, and ship production-grade LLM agents that reason over workforce and skills data on top of Loomra's semantic layer. You won't just prototype — you'll own agent workflows end to end, from design through evaluation, deployment into the tools employees use daily (Teams, Slack, Copilot), and iteration in front of enterprise customers. You'll work alongside product, data, and platform engineers to turn powerful capabilities into reliable, safe, and fast product experiences.
If you've built agents that actually made it to production — and you care as much about evaluation, guardrails, and reliability as you do about capability — we want to talk to you.
Responsibilities
- Design and build multi-agent systems and orchestration - intent routing, planning, tool use, and coordination across specialized agents.
- Implement retrieval and RAG pipelines over structured and unstructured workforce data, grounded in our knowledge graph connecting people, jobs, and skills.
- Integrate LLMs with tool/function calling and protocols such as MCP to give agents controlled access to HCM systems, business logic, and workflows.
- Build evaluation harnesses, guardrails, and safety/bias checks, and work within the governance engine so agents behave reliably, respect customer policies, and produce a full audit trail.
- Ship agents in a model-agnostic way across providers (Anthropic, Google, IBM watsonx) and deploy them into Teams, Slack, and Copilot.
- Optimize agents for latency, cost, and reliability at enterprise scale.
- Take agents from prototype to production — with monitoring, observability, and a fast iteration loop.
- Partner closely with product, data, and platform teams to translate customer needs into agent capabilities.
Requirements
- 5+ years building production software
- Proven experience building and shipping LLM agents to production — not just demos or prototypes.
- Hands-on with at least one agent orchestration framework (e.g. LangGraph, LangChain, AutoGen, CrewAI, Semantic Kernel, or similar).