Machine Learning Engineer

Engineer the Next Generation of AI Models

About Tzafon

Tzafon is on a mission to become the world's leading lab in multi-agent coordination. We are dedicated to developing steerable and trustworthy collective systems by creating a groundbreaking form of collective intelligence that seamlessly integrates humans and AI.

From our twin hubs in Tel Aviv and San Francisco, a hand-picked crew of mathematicians, researchers and repeat founders work as peers, move fast, and own everything they touch.

Our single goal: become the world’s reference lab for large-action modelling and multi-agent coordination at scale.

If you're driven by innovation and excellence, and eager to contribute to groundbreaking work in multi-agent coordination, Tzafon is the place for you.

What You’ll Do

• Design and train large-action models (LAMs) that take sophisticated, multi-step actions across full operating systems and distributed agent swarms.
• Run colossal training jobs—orchestrate thousands of GPUs, optimize throughput, and squeeze every last FLOP from modern hardware.
• Invent new methods that marry game-theoretic RL and OS-level action spaces—and turn them into production systems.
• Build the infra: data pipelines, simulation environments, evaluation harnesses, and rollout schedulers that keep petaflop clusters humming.
• Collaborate ultra-tightly with a tiny, senior team who live by the mantra ships-not-slides.

You’re a Great Fit If You Have

• Deep mastery of ML/NLP/LLMs, plus real-world experience pushing models past the billion-parameter mark.
• Fluent Python; Rust is a bonus for high-performance infra.
• HPC & distributed-training skills—including troubleshooting NCCL issues, implementing sharded optimizers, managing cluster resources, and performance profiling at the kernel level.
• Serious maths chops—with applied skills in game theoretic algorithms and discrete optimization.
• Demonstrated success with multi-agent or OS-action models, robotic control, or autonomous software agents.

Extra Credit

• Papers at NeurIPS / ICML / ICLR on RL, LLM tool-use, or large-action modelling.
• Founder or early engineer at a successful AI / infrastructure startup.
• Built or maintained 1K+-GPU training clusters.
• Major OSS contributions adopted across the community.
• Original research in game theory or discrete mathematics.

Location & Lifestyle

• On-site in Tel Aviv at Rooms Neve Tzedek—an energetic neighborhood that sparks ideas.
• Full-time (40 hrs/week).

Compensation & Perks

• Salary $100k – $150k (gross, DOE). As Tzafon scales, so will your upside.
• Equity Generous equity package.
• Growth Budget for top conferences & continuous learning.
• Gear The best tools—A100/H100 time, dev boxes.
• Culture Tiny, elite, ownership-driven team.
• Relocation Visa & relocation assistance to Tel Aviv.