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): @Clyde_Klein
Email (Optional) : [email protected]
