Software Dev / Architect - Simulation /AI (DevOp / MLOp) at Humanitas Solutions
Montréal, QC, Canada -
Full Time


Start Date

Immediate

Expiry Date

02 Sep, 25

Salary

55000.0

Posted On

02 Jun, 25

Experience

5 year(s) or above

Remote Job

Yes

Telecommute

Yes

Sponsor Visa

No

Skills

Computational Physics, Storage Systems, Hardware Diagnostics, Reinforcement Learning, Computer Engineering, Robotics, Version Control, Physics Engines, French, Rust, Unreal Engine, Python, Pipelines, Multi Agent Systems, English, Simulations, Docker, C++, Containerization

Industry

Information Technology/IT

Description

THE COMPANY

Humanitas is a young, award-winning innovator startup based in Montreal, specializing in emergency response and resilient technologies. Working with a list of world-class industry leaders and researchers, our team specializes in advanced telecom, simulation, visualization, cybersecurity, swarming robotics, edge computing, and more. Our multidisciplinary team also endeavors to universalize our technology and expand their applications to routine use cases beyond edge scenarios.
We are an ambitious group of young people who aim to contribute to a little change in the world by creating IT solutions that help people globally, especially when they need it most. Compassion is at the core of our business, and our collaboration is driven by our desire to challenge our limits and explore our potential.

EDUCATION:

  • Master’s degree required, PhD preferred in Computer Science, Computer Engineering, Computational Physics, Simulation Engineering, or a related field.

EXPERIENCE:

  • 5+ years designing and implementing high-fidelity simulation environments for robotics, multi-agent systems, game development, or synthetic data pipelines.
  • Proven expertise with Unreal Engine, or similar platforms for building 3D interactive environments.
  • Experience integrating simulations with ML, LLM, or reinforcement learning pipelines.

TECHNICAL SKILLS:

  • Proficiency in C++, Python, or Rust
  • Strong grasp of 3D graphics, physics engines, and animation systems
  • Familiarity with cloud-based rendering and scalable compute (AWS/GCP/Azure)
  • Experience in containerization (Docker, Kubernetes) and version control
  • Strong experience in configuring and managing hardware systems (e.g., GPU servers, high-throughput storage systems, edge compute devices)
  • Comfortable with systems monitoring, hardware diagnostics, and server-level performance profiling
  • Bonus: Knowledge of robotics simulators (Gazebo, Isaac Sim), RL frameworks (OpenAI Gym, PettingZoo), or LLM fine-tuning pipelines
Responsibilities

THE ROLE: SENIOR SIMULATION SYSTEMS ARCHITECT

We are seeking a Simulation Systems Architect with deep expertise in designing and integrating advanced simulation ecosystems for training, testing, continuous learning, and deployment in complex, real-world environments.
This role is essential to the development of our comprehensive simulation pipeline — a platform enabling large-scale, synthetic data generation, real-time system testing, continuous model feedback, and automated deployment of decision-making systems. It supports applications spanning robotics, language interaction, perception, and federated or distributed systems. This role also includes responsibility for ensuring infrastructure scalability and performance, including simulation server configuration and hardware optimization.

KEY RESPONSIBILITIES

  • Simulation Architecture & Design

  • Lead the design of a modular, scalable simulation platform using engines like Unreal Engine, supporting multimodal synthetic data generation (vision, audio, text, tabular).

  • Architect interoperability between simulation outputs and machine learning, analytics, or real-time feedback pipelines.
  • Define data standards, simulator APIs, and interfaces for deploying agents, collecting ground-truth metadata, and embedding domain randomization techniques.

  • Pipeline Implementation

  • Oversee end-to-end pipelines from synthetic data generation through data cleaning, normalization, storage, and integration with downstream processing (model training, LLM fine-tuning, etc.).

  • Build support for continuous learning loops — integrating simulator-based data with model feedback and real-world telemetry to guide retraining.
  • Lead efforts to integrate simulation data with vector databases and retrieval systems to support hybrid learning applications (e.g., retrieval-augmented generation).
  • Configure local and cloud-based simulation servers to ensure sufficient memory, GPU availability, and I/O throughput for synthetic data rendering and real-time agent feedback.

  • Deployment & Infrastructure

  • Collaborate with DevOps and MLOps teams to deploy simulation environments across cloud, edge, and on-premise environments.

  • Ensure simulation environments are containerized, repeatable, and version-controlled.
  • Support deployment-ready packages for simulations used in testing perception systems, conversational agents, or autonomous robotics.
  • Design and optimize high-performance simulation clusters, including bare-metal servers, edge nodes, and containerized environments.
  • Select and tune hardware (CPU, GPU, memory, network) for specific simulation workloads (e.g., rendering, inference, or data streaming).

  • Benchmarking & Evaluation

  • Establish validation metrics and test harnesses for evaluating system performance under simulated and real-world scenarios.

  • Implement benchmarking standards for agents across RL, supervised learning, and fine-tuned LLMs within simulated tasks.

  • Cross-Team Integration

  • Work closely with ML researchers, robotics engineers, data scientists, and embedded developers to ensure simulations match deployment constraints and research needs.

  • Provide technical mentorship and lead architectural reviews for simulation-related initiatives.
Loading...