Runs the team day to day
In-house, owns the opening
Joined the team recently
From your own Gmail and LinkedIn.
About the role
This is a hands-on software architect role in Tel Aviv, listed as remote, for an engineer with at least five years as an architect or principal engineer who still writes production code. You will design and build the generative AI systems behind healthcare AI agents that answer patient questions, book appointments and handle prescription refills. The employer sells AI agent software to health systems for patient communications across calls, web, SMS and apps.
What you’ll do
- Own end-to-end architecture across several teams, writing design docs and setting technical standards.
- Choose the model strategy, weighing small against large models, self-hosted against API models, fine-tuning, and routing between providers on quality, latency and cost.
- Turn product prototypes into platform capabilities without a full rewrite.
- Design serving infrastructure for small and large models that meets strict latency and availability targets.
- Build anomaly detection and classification systems that catch failures in live conversations.
- Deliver voice and other channel experiences with low latency and high accuracy.
- Lead new AI initiatives, including complex features and third-party integrations.
- Add guardrails, hallucination controls and explainability for healthcare-grade agents.
- Write production code, prototype proofs of concept, and set review standards for core agent services.
What they’re looking for
- Five or more years as an architect or principal engineer.
- Two or more years building ML or LLM applications in production.
- A bachelor's or master's degree in computer science, data science or a related field.
- Hands-on Python skills, with production code and prototypes as part of the work.
- Deep production experience with LLM systems, including context engineering, RAG, tool calling, structured outputs and agent frameworks such as LangGraph.
- Experience designing evaluation frameworks for non-deterministic systems and using them to make shipping decisions.
- Experience with real-time, latency-sensitive systems such as streaming, WebSockets, WebRTC or event-driven architectures.
- Comfort moving between fast 0-to-1 builds and hardening platforms at scale.
- Strong leadership, interpersonal and communication skills in a fast-changing environment.
Nice to have
- Knowledge of MLOps.
- Voice AI experience with STT, TTS, speech-to-speech models or LiveKit.
- Model serving and optimization with vLLM or Triton, quantization, GPU efficiency or fine-tuning small models.
- LLM observability tooling such as Langfuse.
- Node.js or TypeScript experience.
What’s on offer
- Unlimited PTO and paid sick leave.
- Company-wide long weekends every quarter.
- Hybrid work with a dog-friendly office in Tel Aviv, plus up to two weeks per year working from anywhere.
- Pension, social benefits and a Keren Hishtalmut study fund.
- A monthly meal allowance.
Skills
Questions about this role
Is this role remote?
It is listed as remote from Tel Aviv. There is also a hybrid office in Tel Aviv, and you can work from anywhere for up to two weeks per year.
How many years of experience do I need?
At least five years as an architect or principal engineer, plus two years building ML or LLM applications in production.
Do I need a degree?
A bachelor's or master's degree in computer science, data science or a related field is listed as a requirement.
What benefits are offered?
Unlimited PTO, paid sick leave, quarterly company-wide long weekends, pension, Keren Hishtalmut, and a monthly meal allowance.