Staff/Principal AI Engineer — Agentic Trading Systems

Описание вакансии

Seeking or Hiring: *✅ Hiring ✅*

Website: https://o1.exchange/

Twitter: https://x.com/o1_exchange

1 Line Description of the Firm (in bullets):

o1.exchange is Onchain Robinhood, where you can build advanced crypto-native strategies over any assets, spot/perp/prediction markets. Join a team of hackers to build the next gen financial platform.

Who we're looking for and what you'll be doing (5 Bullet points max):

Staff/Principal AI Engineer — Agentic Trading Systems

We're hiring a Staff/Principal AI Engineer to design and build the agentic models at the core of our trading platform. You'll own the full lifecycle — from model architecture and fine-tuning through inference, deployment, and live performance — building autonomous agents that ingest market data, reason over it, and execute trading strategies under real capital and real latency constraints. This is a hands-on senior role: you set technical direction, write production code, and are accountable for how the models perform in the market.

Responsibilities

\* Design and build agentic trading models with planning, reasoning, tool use, memory, and execution across market data and trading venues.

\* Own the full model lifecycle, including data pipelines, training, fine-tuning, evaluation, inference serving, monitoring, and continuous improvement.

\* Apply full fine-tuning, LoRA / PEFT, RLHF / RLAIF, and reward modeling to improve trading-specific reasoning and decision-making.

\* Optimize inference for low latency and high throughput through quantization, batching, caching, distillation, and serving infrastructure.

\* Build backtesting and simulation environments to evaluate agent behavior before deploying real capital.

\* Develop guardrails, risk controls, and observability systems to ensure safe and predictable production behavior.

\* Set technical standards and mentor engineers across AI and quantitative teams.

Requirements

\* 7+ years of experience building and shipping production ML / AI systems, with hands-on ownership across the full model lifecycle.

\* Strong expertise in model tuning, PEFT, reinforcement learning methods, and inference optimization across latency, throughput, cost, and serving.

\* Experience building agentic systems involving planning, tool use, memory, multi-step reasoning, and orchestration.

\* Strong Python skills and experience with PyTorch, Hugging Face, vLLM, TensorRT, or similar technologies.

\* Solid software and systems engineering fundamentals, with experience handling large-scale, real-time, and time-series data.

\* A rigorous, evaluation-driven mindset with strong backtesting, simulation, and validation practices.

Nice to Have

\* Experience in trading, quantitative finance, digital assets, or other latency- and risk-sensitive domains.

\* Background in reinforcement learning for sequential decision-making.

\* Experience operating models under strict reliability, risk, or regulatory constraints.

Telegram (person filling the form): Откликнуться

Email (Optional) : Откликнуться