Description Overview
Were seeking a GenAI Solutions Architect who can design and own the architecture for LLM and agent-based solutions - from problem framing and system design to implementation and go-live. Youll work closely with product, customers, and engineering teams to define agentic workflows, select the right models and infrastructure, and ensure our GenAI solutions are reliable, safe, and scalable in production.
What Youll Do
- Design end-to-end GenAI and agentic architectures (multi-agent workflows, tools, memory, RAG pipelines) for product and enterprise use cases
- Translate business problems into GenAI solution blueprints, including model selection, orchestration, data flows, and integration points
- Define standards and best practices for prompt design, tool-calling, context management, and long-running workflows
- Architect retrieval and knowledge pipelines (RAG, vector stores, embeddings, indexing strategies)
- Evaluate and recommend LLMs/model providers (OpenAI, Azure OpenAI, AWS Bedrock, open-source models, etc.) based on use case, cost, and constraints
- Establish evaluation strategies for GenAI solutions (offline/online evals, guardrails, safety checks, quality metrics, human-in-the-loop feedback)
- Work with engineering to implement agent-based systems using Lyzrs platform and modern backend stacks (Python preferred)
- Define APIs, contracts, and integration patterns between agents, microservices, and external systems
- Drive non-functional requirements for GenAI systems: latency, scalability, observability, cost
optimization, and resilience
- Partner with product managers and customers to shape requirements and define solution scope, risks, and trade-offs
- Prepare architecture diagrams, decision records, and technical documentation for internal and external stakeholders
- Ensure solutions comply with security, privacy, and compliance requirements for AI systems (data handling, PII, access control)
- Define monitoring and observability strategies for GenAI agents (logging, tracing, hallucination detection, drift monitoring)
- Continuously refine architectures based on production feedback, performance data, and new GenAI capabilities.
What Were Looking For
- 6+ years of experience in software/solution architecture or backend engineering, with at least 12 years hands-on with LLM/GenAI systems
- Practical experience designing and delivering GenAI solutions (e.g., RAG systems, chat/agents, document copilots, workflow understanding of :
- LLMs, embeddings, prompt engineering, and tool-calling
- Vector databases (Qdrant, Weaviate, Chroma, PGVector) and retrieval patterns
- System design and API architecture
- Experience with at least one major cloud (AWS required; Azure or GCP is a plus)
- Ability to balance experimentation with production rigor (cost, performance, reliability, safety)
- Strong ownership, product mindset, and comfort working directly with customers and cross-functional teams
- Excellent communication skills to simplify complex GenAI concepts for technical and non-technical stakeholders
Nice To Have
- Experience working in fast-paced startup environments
- Exposure to AI/ML product ecosystems
(ref:hirist.tech)
