Staff Engineer, Autonomous Flight Modeling

Merlinlabs; Boston (MA); On-site; USD 160,000 - 230,000; Full time

Benefits offered by this job

Catered lunches

Snacks

Beverages

Health, dental, life insurance

Unlimited vacation

401k with match

Job summary

Merlin Labs is seeking a senior researcher to design and evaluate world models predicting aircraft state and environment. You will build rollout and imagination tooling to compare candidate plans against predicted outcomes and own calibrated uncertainty signals for safety monitoring.

You will work on validating models against flight data, exploring the sim-to-real gap, and benchmarking against physics-based models.

Qualifications

• 10+ years building and training deep learning models, with sequence/dynamics/trajectory focus.
• Strong PyTorch and training stack knowledge: distributed training, mixed precision, experiment tracking.
• Reading and understanding flight dynamics models and what the network replaces.
• Demonstrated rigor in evaluation and uncertainty quantification.
• Comfort with messy real-world sensor/telemetry data.

Responsibilities

• Design, train and evaluate world models predicting aircraft state and environment.
• Build rollout/imagination machinery for evaluating candidate plans.
• Own calibrated predictive uncertainty and out-of-distribution signals for safety monitoring.
• Characterize validity boundaries across flight regimes, weather and traffic densities.
• Collaborate on sim-to-real with Data/Sim team: train in simulation, validate with flight data.
• Benchmark learned dynamics against physics-based models and flight-controls; report on where classical approaches win.
• Contribute to the shared eval harness with reproducibility (data version, config, seed).

Skills

Deep learning

Sequence modeling

Uncertainty quantification

PyTorch

Distributed training

Experiment tracking

Flight dynamics

State estimation

Education

Degree in Computer Science/AI/Data Science/Computer Engineering/Applied Math

Job description

Merlin Labs is seeking a senior researcher to design and evaluate world models predicting aircraft state and environment. You will build rollout and imagination tooling to compare candidate plans against predicted outcomes and own calibrated uncertainty signals for safety monitoring.

You will work on validating models against flight data, exploring the sim-to-real gap, and benchmarking against physics-based models.