Senior AI/ML Engineer - LLM Orchestration & Real-Time AI

Webigo AI; United States; Remote; USD 150,000 - 195,000; Full time

Benefits offered by this job

Home office stipend

Comprehensive medical, dental, and vis

Annual learning allowance

Unlimited PTO with 3 weeks minimum

Equity grants

Job summary

Webigo AI is seeking a senior principal engineer to drive the architectural evolution of an autonomous agent platform. You will own end-to-end production AI systems, from design through benchmarking to ship, focusing on LLM orchestration, multi-tenant vector memory, and real-time scheduling.

The role is 100% remote, globally open, with a home office stipend and an annual learning allowance. Strong collaboration with frontend teams will ensure sub-second streaming experiences and scalable

Qualifications

• 5+ years of professional engineering experience in production AI systems.
• Strong Python/TypeScript coding skills and production-grade software practices.
• Hands-on experience with LLM orchestration frameworks.

Responsibilities

• Architect low-latency agent loops integrating LLMs, function calling, and structured outputs.
• Design multi-tenant vector memory and caching for scalable workloads.
• Collaborate with frontend engineers to deliver sub-second streaming UI experiences.
• Benchmark model accuracy, safety guardrails, and latency across thousands of tenants.
• Own reliability and cost profile of production AI services end-to-end.

Tools

Python

TypeScript

LangChain

LlamaIndex

Pinecone

pgvector

Qdrant

Job description

Webigo AI is seeking a senior principal engineer to drive the architectural evolution of an autonomous agent platform. You will own end-to-end production AI systems, from design through benchmarking to ship, focusing on LLM orchestration, multi-tenant vector memory, and real-time scheduling.

The role is 100% remote, globally open, with a home office stipend and an annual learning allowance. Strong collaboration with frontend teams will ensure sub-second streaming experiences and scalable