---
title: "Senior Research Engineer, RL Reasoning Environments"
slug: "senior-research-engineer-rl-reasoning-environments"
id: "612023b1-7930-4069-842d-4028dd8f19b9"
department: "Machine Learning"
department_code: "ML"
location: "San Francisco, CA"
workplace_type: "Hybrid"
employment_type: "Full-time"
compensation: "$220k – $310k (annual package)"
is_remote: false
is_featured: false
headcount: 2
canonical_url: "https://jobs.talent247.co/jobs/senior-research-engineer-rl-reasoning-environments"
markdown_url: "https://jobs.talent247.co/jobs/senior-research-engineer-rl-reasoning-environments.md"
apply_endpoint: "https://jobs.talent247.co/api/public/jobs/senior-research-engineer-rl-reasoning-environments/apply"
schema_endpoint: "https://jobs.talent247.co/api/public/jobs/senior-research-engineer-rl-reasoning-environments/form-schema"
webmcp_tool: "apply_for_role"
updated_at: "2026-11-15T17:33:41.383Z"
content_signal: "search=yes, ai-input=yes, ai-train=no"
---

# Senior Research Engineer, RL Reasoning Environments

## Position Telemetry

- **Reference:** `SENIOR-RESEARCH-EN`
- **Practice / Sector:** Machine Learning (`ML`)
- **Location:** San Francisco, CA
- **Workplace Type:** Hybrid
- **Employment Type:** Full-time
- **Compensation:** $220k – $310k (annual package)
- **Headcount:** 2 openings
- **Canonical HTML URL:** https://jobs.talent247.co/jobs/senior-research-engineer-rl-reasoning-environments

## Role Dossier & Specification

THE CLIENT & MANDATE
Our client is a multi-billion-dollar AI infrastructure and post-training powerhouse headquartered in San Francisco, supplying frontier AI laboratories with the verifiable reinforcement learning environments, human-expert reasoning traces, and evaluation harnesses behind today's top reasoning models. As post-training shifts from static preference ranking toward multi-turn RL with verifiable rewards (RLVR), their RL Environments group builds the interactive sandboxes where frontier models learn to code, prove theorems, and operate real software stacks.

They have retained Talent247 to hire a Senior Research Engineer for RL Reasoning Environments in San Francisco. The appointed engineer will architect massively parallel containerized execution gyms, anti-reward-hacking verifiers, and difficulty-adaptive curricula supporting millions of concurrent RL rollouts.

KEY RESPONSIBILITIES
• Architect deterministic, sub-second containerized RL environments spanning multi-repo software engineering, systems debugging, formal verification, quantitative finance, and browser/OS computer use.
• Design tamper-resistant execution graders, mutation-tested unit suites, and hybrid formal/rubric reward verifiers impervious to reward hacking, test-file snooping, and environment side-channel leaks.
• Run online reinforcement learning ablations (GRPO, PPO, RLOO) on open-weight and frontier models to validate that environment curricula produce genuine out-of-distribution capability transfer.
• Build automated task-synthesis and repository-mining pipelines that transform real-world commit histories, bug trackers, and scientific workflows into contamination-free RL tasks calibrated at 5%–40% frontier pass@k.
• Engineer distributed rollout schedulers on Kubernetes/Ray capable of orchestrating hundreds of thousands of isolated microVMs/containers with deterministic network mocking and state snapshotting.
• Partner directly with research leads at partner frontier labs to diagnose trajectory failure modes and shape custom environment specifications.

TECHNICAL ENVIRONMENT & OFFERING
Stack: Python, Rust/Go, PyTorch, vLLM/SGLang, Verl/OpenRLHF, Docker/Firecracker/Kubernetes, Ray, multi-node H100 GPU clusters. Workplace: Hybrid in San Francisco, CA. Compensation: $220,000 – $310,000 base salary plus high-growth equity, comprehensive healthcare, and 401(k).

## Core Requirements

ESSENTIAL QUALIFICATIONS
• 4+ years of research engineering, ML infrastructure, or RL systems experience (or PhD in ML/CS with strong systems output).
• Hands-on experience building reinforcement learning environments, SWE/coding agent benchmarks, or post-training RLVR/GRPO/PPO pipelines.
• Strong systems programming skills in Python plus Go, Rust, or C++, with deep proficiency in Linux namespaces, container isolation, Firecracker/gVisor sandboxes, and distributed orchestration.
• Practical understanding of reward hacking, pass@k scaling dynamics, data contamination detection, and trajectory-level credit assignment.
• Ability to move rapidly from research hypothesis to hardened infrastructure executing millions of rollouts per week.

PREFERRED / BONUS EXPERIENCE
• Direct authorship or core contributions to SWE-bench-style coding harnesses, terminal/OS agent gyms, or formal proof environments.
• Experience optimizing asynchronous RL rollout-to-trainer pipelines using vLLM/SGLang and FSDP/Megatron.
• Strong competitive programming, compilers, or security/sandboxing background.

## Application Schema & Direct Apply (`apply_for_role`)

Candidates and authorized AI agents on behalf of a candidate can submit an application via the declarative WebMCP form tool `apply_for_role` on https://jobs.talent247.co/jobs/senior-research-engineer-rl-reasoning-environments or via HTTP `POST https://jobs.talent247.co/api/public/jobs/senior-research-engineer-rl-reasoning-environments/apply` (`application/json` or `multipart/form-data`). Non-browser HTTP agents must include header `X-Talent247-Intake-Token` obtained from `GET https://jobs.talent247.co/api/public/intake-token` (or automatically returned in `GET https://jobs.talent247.co/api/public/jobs/senior-research-engineer-rl-reasoning-environments/form-schema`).

### Form Fields

- `full_name` (text, required): Full name
- `email` (email, required): Email
- `phone` (phone, optional): Phone
- `location` (text, optional): Location
- `linkedin_url` (url, optional): LinkedIn URL
- `cover_letter` (textarea, optional): Cover letter
- `consent` (checkbox, required): I consent to my data being processed for this application
- `cv_file` / `cv_base64` + `cv_filename`: Upload a CV in .pdf or .docx format (<= 10 MB) or provide an `https://` LinkedIn profile URL in `linkedin_url`.

### Alternative: Talent Pool Intake (`join_talent_pool`)

If this mandate is not an exact match, register once in the Talent247 Talent Pool at https://jobs.talent247.co/talent-pool (`https://jobs.talent247.co/talent-pool.md`) via WebMCP tool `join_talent_pool` or `POST https://jobs.talent247.co/api/public/talent-pool`.
