- Job title
- Member of Technical Staff (Research Engineering)
- Job location
- New York City
- Job description
- # About us
**ATG (Autonomous Technologies Group)** is an AI lab deploying frontier reasoning systems within financial markets.
**Founders:** Early GPU cloud (9 figure exit).
**Investors:** Garry Tan / YC + Founder of one of the most successful quant funds, BoxGroup (Plaid, Ramp, Stripe), top-tier angels.
**About the Role**
You’ll bridge research and engineering—rapidly implementing, experimenting with, and scaling new algorithms and models. You’ll work closely with scientists and founders to translate ideas into high-performance systems, and will operate across the stack from prototyping to deployment.
**Responsibilities**
- Prototype and scale experimental models (LLMs, RL agents, agentic systems) on large, real-world data.
- Build tools and pipelines for training, evaluation, and analysis.
- Implement state-of-the-art research from papers and iterate in collaboration with scientists.
- Own the full ML lifecycle: data engineering, experimentation, training, and deployment.
- Operate in a highly autonomous, engineering-driven environment.
**Requirements**
- Strong background in deep learning, reinforcement learning, or computational modeling.
- Expert-level Python and significant experience in ML frameworks (PyTorch, JAX, TensorFlow).
- Ability to translate research into robust, scalable systems.
- Experience with distributed computing or large-scale ML pipelines.
- Demonstrated curiosity and hands-on experimentation skills.
- **Expertise in leveraging the latest AI tools (Cursor, Claude Code, Codex, etc) to increase productivity & code output while maintaining high code quality, maintainability, and structure.**
**Why Join ATG**
We're building a small, elite team. If you're excited by AI, markets, and building from first principles, we’d love to meet you.
- Work on AI with a massive market opportunity
- Early team of repeat founders backed by top investors
- High agency, talent dense, zero bureaucracy
