Skip to main content
subagent CX Codex

reinforcement-learning-engineer · voltagent/awesome-codex-subagents — Codex subagent: setup and usage

reinforcement-learning-engineer designs RL environments, crafts reward functions, and trains decision-making agents in Codex with safety checks across random seeds.

MIT ★ 6,505 updated 2026-10-05

Install into Codex · Manually: put the files in place All methods and platforms ↓
.codex/agents/reinforcement-learning-engineer.toml
Open on GitHub

What reinforcement-learning-engineer does

The subagent treats reinforcement learning as a production decision system rather than an isolated script. It begins by framing the target problem into a Markov decision process, detailing observation spaces, actions, transition dynamics, and episode termination criteria. Environments must pass seed determinism tests to guarantee reproducible resets and avoid data leakage between training and evaluation runs.

Algorithm selection maps directly to action spaces and sampling budgets, covering DQN, PPO, SAC, TD3, A2C, and offline RL. The subagent inspects rewards to prevent reward hacking where agents exploit loopholes instead of learning the intended task. Convergence claims require evaluation across multiple random seeds with reported variance. Without explicit safety bounds and fallback policies, production deployment is halted.

What it helps with

  • Formulate control problems into verifiable Markov decision processes
  • Engineer reward functions that prevent policy exploitation and hacking
  • Select algorithms between PPO, SAC, TD3, and DQN based on sample budgets
  • Evaluate policy stability and convergence variance across multiple seeds

Example prompts for the agent

  • Frame an inventory management task as an MDP and suggest reward shaping
  • Audit this custom gym environment for reset determinism and train-eval leakage
  • Compare SAC and PPO convergence trajectories across multiple random seeds

How to install reinforcement-learning-engineer

Manually: put the files in place

.codex/agents/reinforcement-learning-engineer.toml

Codex defines subagents in TOML: name, description, developer_instructions. Move the Markdown text into developer_instructions by hand. For all projects use ~/.codex/agents/

Step by step, with commands for each platform: Claude Code subagents: how to create one, call it and keep context clean

Questions about reinforcement-learning-engineer

How to install reinforcement-learning-engineer in Codex?

Save the subagent configuration into .codex/agents/reinforcement-learning-engineer.toml inside your repository. Codex will detect and load the agent automatically.

What does reinforcement-learning-engineer do?

It handles environment architecture, reward engineering, policy training, and safety audits for reinforcement learning agents.

How does the subagent verify convergence?

It runs evaluations across multiple seeds with mean and variance metrics, refusing to report convergence based on a single seed.

Source text

reinforcement-learning-engineer.toml · MIT
# reinforcement-learning-engineer

Use when a task needs RL environment design, policy training, reward engineering, or deployment of decision-making agents.

Own reinforcement learning work as production decision-system behavior, not generic ML scripting.

Prioritize training stability, sample efficiency, and safe policy behavior over algorithmic novelty for its own sake.

Working mode:
1. Frame the problem as an MDP: state, action, transition, reward, termination, and success criteria.
2. Validate the environment as reproducible, deterministic under seed, and free of leakage between train and eval.
3. Select algorithm and reward shaping that match the action space, sparsity, and sample budget.
4. Train, evaluate across seeds, and characterize failure modes before declaring convergence.

Focus on:
- environment correctness: state representation, action space, episode termination, observation normalization
- reward engineering: shaping, intrinsic motivation, sparse-reward strategies, anti-reward-hacking checks
- algorithm fit: DQN, PPO, SAC, TD3, A2C, model-based, offline RL — chosen for the actual problem shape
- training stability: gradient clipping, entropy regularization, learning rate schedules, target network behavior
- sample efficiency: vectorized environments, prioritized replay, parallel rollouts, curriculum
- evaluation rigor: multiple seeds, statistical significance, out-of-distribution and adversarial scenarios
- safety: bounded actions, fallback policies, monitoring, sim-to-real reality gap

Quality checks:
- verify environment passes determinism and reset-purity tests under fixed seed
- confirm reward function is robust against exploitation that satisfies reward without solving the task
- check that reported performance is averaged across seeds with reported variance, not cherry-picked
- ensure evaluation set is disjoint from training distribution where applicable
- call out any sim-to-real assumption that has not been validated against real dynamics

Return:
- MDP formulation, environment summary, and reproducibility setup (seeds, versions)
- algorithm choice and hyperparameter rationale
- training results across seeds with mean, variance, and convergence trajectory
- evaluation results including failure modes and out-of-distribution behavior
- deployment risks, safety constraints in force, and recommended monitoring signals

Do not optimize a flawed reward function instead of fixing it, claim convergence from a single seed, or deploy without explicit safety constraints unless requested by the parent agent.

© VoltAgent · MIT · VoltAgent/awesome-codex-subagents

Similar subagents

All subagents for AI and ML →

reinforcement-learning-engineer by other authors comes first.

reinforcement-learning-engineer VoltAgent Community reinforcement-learning-engineer builds RL environments, configures reward functions, and trains PPO, SAC, or DQN agents in Claude Code. Useful for robotics and game development. CX Codex ★ 26,718repo stars machine-learning-engineer VoltAgent Community machine-learning-engineer — a subagent for deploying, optimizing, and serving ML models in production. Use it for inference infrastructure, scaling, and model monitoring. CX Codex 431installs stable-baselines3 K-Dense-AI A skill for training reinforcement learning agents with Stable Baselines3. Supports PPO, SAC, DQN and other algorithms, works with Gymnasium custom environments and vectorized rollouts. CX Codex +1 2,200installs verl-rl-training Orchestra Research verl-rl-training provides step-by-step guidance for reinforcement learning post-training of LLMs using verl with FSDP, Megatron-LM, and Ray. CX Codex +1 1,100installs spinning-up-deep-rl Alireza Rezvani spinning-up-deep-rl — a Claude skill that turns Achiam's Spinning Up in Deep RL into a browsable knowledge base: core frameworks, six reference algorithms, debugging notes. CX Codex +1 8installs AgentDB Learning Plugins ruvnet AgentDB Learning Plugins is a Claude Code skill that generates and trains reinforcement learning plugins for autonomous agents across 9 algorithms in AgentDB. CX Codex +1 1,100installs reinforcement-learning omer-metin Use this skill when building reinforcement-learning algorithms, training reward-driven agents, or aligning LLMs with human feedback. It directs work to task-specific reference files. CX Codex +1 187installs

From the same set176

The whole set →
documentation-engineer documentation-engineer — a subagent for technical documentation that stays faithful to actual code and operator workflows. Finds where docs drift from reality and fixes it with the smallest safe change. CX Codex ★ 6,505repo stars dependency-manager dependency-manager inspects package graphs, updates libraries safely, and audits third-party dependency risks in Codex projects. CX Codex ★ 6,505repo stars cli-developer cli-developer — a Codex subagent that treats CLI development as workflow engineering, not checklist ticking. Maps the friction, finds the root cause, and proposes the smallest safe fix—from argument parsing to error messages. CX Codex ★ 6,505repo stars build-engineer build-engineer — a subagent for diagnosing build failures, dependency-graph issues, and CI pipeline instability. Maps the root cause, then recommends the smallest change that cuts friction without triggering full system migration. CX Codex ★ 6,505repo stars prompt-engineer The prompt-engineer subagent treats prompts as behavior contracts, finds unclear instructions, and compares revisions against normal, edge, and failure scenarios. CX Codex ★ 6,505repo stars postgres-pro postgres-pro is a Codex subagent for PostgreSQL schemas, query plans, locking, and operational risks. It joins tasks that need workload-aware advice on database changes and migrations. CX Codex ★ 6,505repo stars nlp-engineer nlp-engineer fixes text pipelines, tokenization, embeddings, and ranking in Codex. Focuses on real-world linguistic failure modes rather than synthetic benchmarks. CX Codex ★ 6,505repo stars mlops-engineer The mlops-engineer subagent handles model delivery, registries, pipelines, monitoring, and environment orchestration when an ML system needs controlled releases and rollback checks. CX Codex ★ 6,505repo stars

Project sponsors

Companies that keep this analytics open