Engineer

About HapiVet

HapiVet is building the world’s first AI-first veterinary practice management platform. Our mission is to help veterinarians and clinic staff work more efficiently, reduce errors, and improve patient outcomes through intelligent, AI-driven workflows.

We’re looking for an Applied LLM Engineer to help us design and productionize AI agents that can understand natural language and safely execute structured actions across scheduling, billing, inventory, documentation, and compliance modules.

Responsibilities

• Agent Orchestration

Build and maintain pipelines that transform free-text instructions into structured, auditable API calls.

Use frameworks like LangGraph, LangChain,Crew.ai, AutoZen, DSPy, or Model Context Protocol to coordinate multi-step workflows.

• Prompting & Context Management

Design prompt strategies and context-injection mechanisms that enforce safety, compliance, and domain-specific accuracy.

Implement memory and retrieval strategies that improve consistency across tasks.

• Evaluation & Guardrails

Develop test datasets of commands → expected structured outputs.

Implement schema validation, RBAC enforcement, allow/deny lists, and human-in-loop confirmation for sensitive operations.

Track accuracy, cost, latency, and hallucination rates in production.

• Applied ML

Fine-tune lightweight models for intent classification or entity recognition.

Build retrieval-augmented generation (RAG) systems for injecting policy, clinical, and workflow knowledge.

• Collaboration

Work with backend engineers to expose APIs as safe, callable tools.

Partner with UX and product teams to design smooth user interactions and error recovery flows.

Collaborate with veterinary advisors to capture and validate real-world workflow pattern

• Qualifications

2–6 years of experience in applied AI/ML, NLP, or intelligent systems engineering.

Strong proficiency in Python and ML frameworks.

Hands-on experience with LLM orchestration frameworks (LangGraph, LangChain, DSPy, Semantic Kernel, or similar).

Track record of deploying AI systems into production (cloud-native, preferably Azure).

Knowledge of evaluation metrics and techniques for LLM reliability.

Familiarity with multi-tenant SaaS architectures, APIs, and distributed systems.

Bonus: healthcare/enterprise SaaS experience or familiarity with HIPAA/PHI compliance.

• What Success Looks Like (First 6–12 Months)

Deliver AI agents that can reliably handle the top 20–30 most common veterinary workflows.

Achieve >90% accuracy on offline evaluation datasets.

Deploy a safe, guardrailed orchestration pipeline that integrates across multiple backend services.

Scale multi-tenant deployments with strong monitoring, cost control, and performance visibility.

• Why Join Us

Be part of a category-defining AI SaaS platform.

Work at the cutting edge of applied LLMs and orchestration.

Build technology that directly impacts healthcare providers and improves patient safety.

Join a fast-moving team backed by experienced founders and domain experts.