job specification

The Role

You'll join the AI Products squad to build the next generation of LLM-powered applications for the staffing industry. This is a product engineering role, not a research role. You'll ship real AI products used by real recruiters and real clients.

We're not looking for someone who trains models from scratch. We're looking for someone who knows how to build with LLMs — orchestrating agents, designing RAG pipelines, crafting prompt architectures, and shipping applications that solve real business problems. If you've built a chatbot, an agent, or a RAG app that actually went to production (or close to it), we want to talk.

.

What You'll Do

• Design and build agentic systems — multi-step AI workflows that reason, retrieve, and act across multiple tools and data sources
• Build RAG (Retrieval-Augmented Generation) pipelines — connect LLMs to company data through vector databases, semantic search, and context injection
• Craft prompt architectures — design system prompts, few-shot examples, chain-of-thought patterns, and guardrails for production LLM applications
• Integrate with APIs and data sources — consume data from our CDI platform (Microsoft Fabric Lakehouse), ATS (JobDiva), VMS platforms, and internal systems
• Ship production-grade applications — not prototypes. Build with error handling, monitoring, logging, fallback strategies, and user-facing quality
• Collaborate with the Data Platform squad — define what data you need in what format, and work with data engineers to make AI-ready datasets available
• Evaluate and iterate on AI quality — build evaluation frameworks, test edge cases, measure accuracy, and improve outputs systematically
• Stay current with the LLM ecosystem — new models, frameworks, and techniques ship weekly. You'll help evaluate what's worth adopting.

What We're Looking For

Must have (1-3 years of experience):

• Shipped at least one LLM-powered application in a real organization (not just personal projects or tutorials). This could be a chatbot, agent, RAG app, content generator, or similar.
• Strong Python skills — you think in Python, not just write it
• Hands-on experience with LLM orchestration frameworks — LangChain, LlamaIndex, CrewAI, AutoGen, Semantic Kernel, or similar
• Experience building RAG systems — vector databases (Pinecone, Weaviate, Chroma, FAISS), embedding models, retrieval strategies, context window management
• Understanding of prompt engineering at a production level — not just getting good outputs, but building reliable, consistent, testable prompt architectures
• Familiarity with LLM APIs — OpenAI, Anthropic, Azure OpenAI, or open-source model serving (Ollama, vLLM)
• Basic understanding of software engineering practices — Git, API design, error handling, logging
• Good communication in English (daily collaboration with a distributed team)

Nice to have:

• Experience with agent frameworks — multi-agent orchestration, tool use, function calling, planning loops
• Familiarity with fine-tuning LLMs (LoRA, QLoRA, instruction tuning) — not required but valuable for future work
• Experience with evaluation frameworks for LLM outputs (RAGAS, custom metrics, human-in-the-loop evaluation)
• Knowledge of NLP fundamentals — text classification, NER, semantic similarity, embeddings
• Exposure to cloud deployment — Azure, AWS, or GCP for hosting AI services
• Familiarity with staffing, HR tech, or recruiting industry — understanding placement lifecycles, compliance, credentialing
• Experience with Microsoft Fabric , Power BI , or Azure ecosystem

Why This Role

• Build AI products at scale from day one. Our data infrastructure covers millions of candidates, Fortune 500 clients. The problems are real and the impact is immediate.
• Full product ownership. You won't be writing prompt templates in a vacuum. You'll own products end-to-end: architecture, data requirements, prompt design, API integration, quality, and iteration.
• Cutting-edge stack. LLMs, agents, RAG, vector databases, agentic workflows — all in production, not demos.
• Growth trajectory. As the team scales toward an AI Products Squad (2-3 engineers), early joiners will shape the team culture and grow into lead roles.