Research Engineer / Applied Researcher · Careers

The role

We’re looking for strong applied researchers and research engineers to help build our next generation of AI systems, as part of our work on recursive self-improvement: agents that keep learning and improving in real, complex environments. You’ll have large-scale GPU compute, real enterprise environments, and complex tasks and data from leading enterprises, and you’ll work hands-on across training, experimentation, building and iteration.

What you’ll do

Build and iterate on AI agents that keep improving on complex, real-world tasks
Work on recursive self-improvement, agent learning, long-horizon reasoning and LLM post-training
Turn technical ideas into working systems that can be run and validated, fast
Build advanced agent systems, including reasoning, tool use, memory, context management and multi-step workflows
Train, experiment and iterate quickly on large-scale GPU compute
Contribute to post-training, including reinforcement learning, SFT and related methods
Build benchmarks, evaluation pipelines and automated graders
Analyze agent trajectories and failure modes to find the main bottlenecks in models, data and systems
Build and curate high-quality training data, both real interaction data and synthetic data
Work with founding members and customer teams to turn well-defined problems into experiments, prototypes and production systems
Keep improving agents’ accuracy, reliability, latency, token efficiency and end-to-end task completion

What we’re looking for

A bachelor’s, master’s or PhD in CS, AI, ML or a related field
Solid foundations in machine learning and deep learning
Strong hands-on engineering: you can turn a technical idea into a working system quickly
Experience with LLMs, AI agents, reinforcement learning, post-training, evaluation or related projects
Fluency in Python and the mainstream deep learning and LLM ecosystem
The ability to read new AI research quickly, and to implement and validate the methods that matter
The ability to debug independently, analyze failure modes and improve systems through experiments
Strong execution, fast learning and ownership
Clear technical communication with customers and teammates when needed, though the core of this role is building and execution

Nice to have

Hands-on experience building production LLM or agent systems
Experience with reinforcement learning, SFT or other post-training methods
Familiarity with PyTorch, vLLM, SGLang, Ray or similar training and inference infrastructure
Experience with long-horizon agents, coding agents or tool-use systems
Experience with benchmarks, LLM-as-a-judge, reward models or evaluation systems
Experience with synthetic data generation or large-scale data pipelines
Papers, open-source projects, competition results or other evidence of technical ability
Experience at a startup or on a fast-moving research or engineering team

How to apply

Apply with your CV and links to your work. If you prefer, email info@evolocity.ai with the subject “Research Engineer”.

Apply for this role

Equal opportunity

Evolocity is an equal opportunity employer. We welcome applicants of all backgrounds and do not discriminate on the basis of race, color, religion, sex, sexual orientation, gender identity, national origin, age, disability, veteran status, or any other legally protected status.