Hiring for Senior Backend Developer

Job Description:

We build AI-powered SaaS products and agentic systems for clients across hospitality,
healthcare, real estate, and professional services. We are looking for a senior backend engineer
who can own the server side end to end: API services, LLM orchestration, RAG pipelines,
background processing, and the integration layer that ties third-party APIs into agentic workflows.
This is an ownership role. You should be comfortable making architecture decisions and
reasoning about reliability, cost, and scale, not just executing tickets.
What You Will Do
• Design and build production backend services in Node.js and TypeScript with clean API
design, sensible boundaries, and maintainable code.
• Implement server-side LLM features: RAG pipelines, tool-calling agents, structured output,
and streaming responses.
• Orchestrate multi-step AI workflows using the Vercel AI SDK, LangChain.js / LangGraph, or
equivalent frameworks.
• Integrate model providers (Claude, OpenAI, Gemini) with proper retry, fallback, timeout, cost,
and latency handling.
• Build and maintain third-party integrations and OAuth flows across CRMs, communication
APIs, payment providers, and data sources, including handling token and session fragility.
• Work with PostgreSQL and Supabase: schema design, row-level security, migrations, and
query performance.

• Implement vector storage and retrieval using pgvector or comparable systems for
embeddings-based search.
• Build background jobs, queues, and scheduled tasks; handle long-running and asynchronous
AI work reliably.
• Integrate Stripe for billing, subscriptions, and webhook handling.
• Add observability and evaluation for AI behavior using tools such as Langfuse, with attention
to output quality and correctness.
• Review code, mentor mid-level developers, and surface technical risk early.
Required Experience
• 5+ years of backend JavaScript/TypeScript (Node.js) with a strong grasp of asynchronous
patterns, error handling, and API design.
• Demonstrated production experience integrating LLM APIs, not prototypes: RAG, function or
tool calling, or agent systems.
• Solid experience with PostgreSQL or Supabase, including schema design and performance
considerations.
• At least one vector database or embedding workflow shipped to production.
• Experience designing integrations against unreliable external services: rate limits, retries,
idempotency, and webhooks.
• Comfort with authentication and OAuth integration patterns.
• Strong written communication, with the ability to reason about trade-offs and document
decisions.
Nice to Have
• LangGraph or comparable agent framework experience for multi-step, stateful agents.
• Vercel AI SDK in production.
• Python and FastAPI experience for cross-stack work on our services side.
• Docker, AWS, Redis, and queue or background-job experience (BullMQ, Celery, or
equivalent).
• n8n or workflow-automation experience.
• Voice AI exposure such as Retell AI or Vapi.
What We Look For Beyond the Checklist
• You treat LLM output as probabilistic and design around it with validation, guardrails, and
evaluation rather than trusting it blindly.

• You anticipate failure modes early, including token and session fragility, rate limits, cost
blowups, and queue backpressure, and scope them as work rather than discovering them in
production.
• You can take a short, ambiguous brief and turn it into a sensible service architecture without
hand-holding.