Member of Technical Staff
Function Health3 months ago
Remote, United States or Austin, TX, USASenior
Responsibilities
- Architect and build stateful graph-based agent workflows with planning, tool use, and memory.
- Integrate LLMs and multimodal models through structured I/O and function or tool calling.
- Build high-reliability APIs and streaming services for real-time inference, speech, and vision.
- Own production readiness through tracing, logging, metrics, rate limiting, circuit breakers, SLOs, and on-call triage.
- Develop offline and online evaluation pipelines with golden sets, LLM-as-judge evaluation, human rubrics, A/B testing, and regression tests.
- Implement hybrid, vector, and graph retrieval, semantic caching, and long-horizon context management.
- Optimize model routing, prompts, tool selection, quantization, and KV-cache or prefill strategies for cost and latency.
- Partner cross-functionally to translate research into robust production systems and mentor engineers through design and architecture decisions.
Requirements
- At least 1 year building agentic AI systems and 6+ years as a full-stack or ML engineer building production backends or ML systems.
- Production experience with Python, Go, or similar languages and agent orchestration frameworks such as LangGraph, PydanticAI, DSPy, or LlamaIndex.
- Experience integrating frontier LLMs and multimodal models through managed APIs or self-hosted serving.
- Strong API design and backend framework experience, including FastAPI or Flask, and event-driven architectures.
- Experience with PostgreSQL, token streaming, throughput tuning, vector databases, hybrid search, and graph or knowledge storage.
- Experience with production evaluations, LLM-as-judge systems, human-in-the-loop review, rubric design, and CI-integrated regression tests.
- Experience with OpenTelemetry, metrics, structured logs, SLOs, dashboards, Kubernetes, Terraform, Docker, GPU scheduling or autoscaling, AWS or GCP, and GitHub Actions.
- Preferred experience includes real-time multimodal systems, streaming ASR, low-latency TTS, WebRTC, vision pipelines, Graph RAG, multi-hop retrieval, safety governance, regulated data, personalization, consumer-scale AI applications, and WebGPU.
- Clear communication, high ownership, prioritization, execution, clinical precision, and a member-first approach.
Benefits
- Competitive salary and benefits package.
- Flexible working hours and a dynamic environment that encourages creativity and innovation.
- Equal opportunity workplace committed to diversity and inclusion.
Tech Stack
Categories