Lead AI Architect — Real-Time Mahjong Engine & LLM Reasoning

Neurons Lab; Roma; In loco; EUR 95.000 - 150.000; Part-time

Descrizione del lavoro

Neurons Lab is seeking a Hands-on Tech Lead for an AI companion in an online mahjong game. You will design and build the mahjong-playing algorithm with RL, imitation learning, or search-based methods, and integrate an LLM reasoning layer.

You’ll own the end-to-end architecture, ensure a 2-second per-move latency budget, define data interfaces, and collaborate with client engineers and CTO to explain trade-offs.

Competenze

• Hands-on ML/AI engineering at production scale.
• Shipped an AI system inside a live product with hard latency limits.
• 6+ years hands-on ML/AI engineering, with real game AI or sequential decision-making work (RL / MCTS / self-play).

Mansioni

• Design and build the mahjong-playing algorithm: choose and defend the approach (imitation learning on hand histories, RL / self-play, search with MCTS, or a hybrid).
• Own the technical architecture end to end: game model + LLM reasoning layer, valid-action mask, win detection, and API contract with the client’s game bridge.
• Hit the 2-second response budget: design and measure the inference path, batching, and caching; keep a latency buffer for the client-facing number.
• Define data and event names from the client (hand histories, event streams); build training and calibration pipelines.
• Build and run the evaluation harness: measure play strength against reference points and validate explanations.
• Stand up LLM observability with Langfuse (async logging, N+1 batching).
• Take over context from leadership and lead the sprint with the AI engineer; collaborate with client CTO and engineers.
• Watch licensing, engine-bridge capabilities, and multi-rule-set scope risks.

Conoscenze

ML/AI engineering

Production systems

RL / MCTS / self-play

Python for ML

Low-latency inference

LLM integration

Cloud AWS

Technical leadership

Strumenti

Langfuse

Descrizione del lavoro

Neurons Lab is seeking a Hands-on Tech Lead for an AI companion in an online mahjong game. You will design and build the mahjong-playing algorithm with RL, imitation learning, or search-based methods, and integrate an LLM reasoning layer.

You’ll own the end-to-end architecture, ensure a 2-second per-move latency budget, define data interfaces, and collaborate with client engineers and CTO to explain trade-offs.