Nvidia Earth 2 Engineer

GlobalLogic has been engaged in exploring opportunities to implement ML/AL/GenAI-powered applications for multinational industrial conglomerate since 2023, aiming to enhance business efficiency. 

You will be working on projects implying complex data processing pipelines for predictive maintenance anaylitics.
 

Projects tech stack: Kubernetes, Terraform, Helm, LLM, ML, AI, Asyncio, Python, Pandas, HayStack, Azure Blob Storage, Azure DevOps, SQL Alchemy, Docker, Docker Compose, PySpark, PostgreSQL, FastAPI

Requirements:

• 4+ years of professional software engineering (Python-centric), including production services at scale
• 2+ years working with GPU-accelerated AI/ML workloads (training and/or high-throughput inference)
• Proven experience ingesting and processing meteorological/climate datasets (NetCDF/GRIB/Zarr) with xarray/dask
• Track record deploying GPU inference with Triton on Kubernetes (autoscaling, A/B, canary, monitoring)
• Strong system design skills (throughput/latency/SLA thinking) and clean, testable code
• Would be good – Bachelor’s/Master’s in CS, Math, Physics, Atmospheric Science, or equivalent practical experience

Job responsibilities:

You will be engaged into the following activities:

• Design, build, and own Earth-2–backed microservices that deliver real-time weather insights into product workflows
• Operationalize AI models (e.g., FourCastNet/CorrDiff or similar): packaging, optimization (mixed precision, TensorRT), and Triton deployment
• Ingest & curate meteo data pipelines, implement preprocessing (regridding, normalization, gap-fill) with xarray/dask
• Optimize GPU performance end-to-end: memory, throughput, batching, concurrency; profile and remove bottlenecks
• Expose APIs (REST/gRPC) with FastAPI, implement versioning, auth, and rate-limiting; write clear SDK/client wrappers
• Integrate geospatial analytics to align predictions with assets, sites, and routes in energy use-cases
• Harden production logging, metrics, tracing, incidents, cost & capacity planning
• Collaborate with data scientists, product, and platform engineers to prioritize impact and ship incrementally
• Document models, data contracts, runbooks, and validation results; contribute to internal best practices and code reviews

What You’ll Work On:

• Climate-aware risk scoring
• Environment-aware automation
• Temporal & geospatial context layers

Tools

• NVIDIA: Earth-2, Omniverse, Modulus, Triton, CUDA, TensorRT
• Python/AI: PyTorch/JAX, xarray/dask, CuPy/Numba, RAPIDS, MLflow
• Geo/Climate: NetCDF/GRIB/Zarr, CF conventions, GDAL/Rasterio, GeoPandas
• Platform: FastAPI, Docker, Kubernetes, Terraform, Helm, Azure Blob, Azure DevOps, PostgreSQL/PostGIS, PySpark