Website:
zapphire.io
Job details:
We are looking for an AI Software Engineer for a well-funded stealth-mode startup operating at the intersection of artificial intelligence and real estate. The mission is to fundamentally reimagine how properties are discovered, evaluated, and transacted — using autonomous AI agents that work on behalf of buyers, sellers, and investors. Backed by leading investors, they are a small and deeply technical team that moves fast and ships things that matter.
What You'll Build
- Design and develop AI agent pipelines using frameworks such as LangChain, LangGraph, CrewAI, or AutoGen — applied to real-world real estate workflows
- Integrate LLMs (OpenAI, Anthropic, Gemini, open-source models) and engineer robust prompting strategies for production-grade reliability
- Build tool-use and function-calling workflows that let agents take meaningful actions — searching listings, evaluating properties, drafting offers
- Implement retrieval-augmented generation (RAG) pipelines, vector databases (Pinecone, Weaviate, pgvector), and long-term memory systems
- Develop backend services (REST / gRPC) to orchestrate multi-step agentic flows with observability, retries, and graceful error recovery
- Collaborate with product and ML teams to rapidly prototype features and take them from zero to production
- Contribute to architectural decisions as our product scales from early users to the broader market
What We're Looking For
Required
- 2–5 years of software engineering experience with Python as the primary language
- Hands-on experience building or integrating LLM-based systems in a production setting
- Strong grasp of prompt engineering, chain-of-thought reasoning, and agent design patterns
- Experience with at least one agentic or orchestration framework (LangChain, LangGraph, CrewAI, AutoGen, etc.)
- Proficiency in REST API design, async programming, and backend service development
- Familiarity with vector stores or semantic search (Pinecone, Weaviate, pgvector, FAISS)
- Solid CS fundamentals — data structures, algorithms, and system design
Nice to Have
- Experience fine-tuning or applying RLHF to open-source models (LLaMA, Mistral, etc.)
- Knowledge of LLM evaluation frameworks (RAGAS, PromptFoo, custom evals)
- Exposure to cloud infrastructure on AWS / GCP / Azure and containerised deployments
- Prior work in proptech, fintech, or data-heavy domains
Click on Apply to know more.