Platforms Lead

About GalaxEye :

We build the world's first SyncFused OptoSAR satellites and the software backbone that runs them — from TT&C and mission ops to high-throughput geospatial AI pipelines. This role is modeled to match top industry platform leads at companies building operational EO + analytics stacks.

Why this role matters

You'll build the platform that connects satellites → ground → AI: mission control, telemetry & command, high-volume imagery ingestion, COG/GeoParquet storage, GPU-backed model training/inference and customer data delivery — reliable at scale and deployable in cloud
and
on-prem.

What success looks like (first 6–12 months)

• Deliver a vetted reference architecture for Mission & Data Platforms (hybrid cloud + hardened on-prem).
• Ship a CI/CD pipeline, container strategy, and a working dev→staging→prod flow for mission services.
• Provision a baseline on-prem hardware spec (compute, GPU, storage, networking) and a costed procurement plan.
• Deploy a production data ingress pipeline that accepts raw imagery (satellite frames, SAR bursts), produces COGs/GeoParquet/tiles, and feeds model training/inference.
• Hire/mentors first 3–6 platform engineers and codify SRE/hygiene standards (SLOs, runbooks, blameless postmortems).

Responsibilities (hands-on leadership)

• Architecture & Tech Direction
• Define end-to-end platform architecture for Mission Management (TT&C, scheduler, health/telemetry) and Data Management (ingest, catalog, processing, serving).
• Ensure platform
  interoperability with AI/foundation models
  — provide model hosting, feature stores, model-versioning, and low-latency inference endpoints for computer-vision & geospatial workloads.
• Tech Stack & Implementation
• Select and own core stack (see "Preferred tech stack" below).
• Design hybrid deployment patterns: cloud native (GCP/AWS)
  and
  hardened on-prem (bare-metal/K8s/OpenStack/VMware).
• DevOps / SRE / Hygiene
• Build CI/CD, IaC, observability, security, secrets management, and blue/green/rolling deployments.
• Define SLOs/SLIs, runbooks, incident response and capacity planning.
• Data & AI Integration
• Work with Data Science/ML engineers to design training pipelines (data versioning, dataset formats such as COG/GeoParquet/Iceberg), GPU scheduling, and model deployment (Triton/KFServing/MLflow).
• Ground & Mission Ops
• Integrate with ground station stacks, scheduling systems, and mission planners. Implement telemetry pipelines (CCSDS-like framing, TM/TC handlers) and mission workflows.
• Hardware Sizing & Procurement
• Size compute/GPU/storage/network (IOPS, throughput, retention) and produce costed specs for procurement. Recommend RAID/erasure/cluster configs (Ceph/Lustre/NAS).
• Team Building
• Recruit, grow, and lead a cross-functional team (Platform Engineers, DevOps, SRE, Systems Administrators). Set OKRs and technical KPIs.

Must-have (non-negotiable)

• 8+ years building production platforms / infrastructure / DevOps (or equivalent).
• Deep, hands-on experience with
  Kubernetes + containers
  , container networking, and cluster ops at scale.
• Proven track record of designing hybrid cloud + on-prem architectures and running secure, high-availability services.
• Strong systems programming / backend experience (Python and/or Go).
• Experience sizing and managing
  GPU clusters
  and high-throughput storage for large raster datasets.
• Solid understanding of geospatial data formats and pipelines (COG, GeoTIFF, GeoParquet, PostGIS) and large raster handling.
• Experience with CI/CD, IaC (Terraform/Ansible), message streaming (Kafka), and orchestration (Argo/Kubernetes).
• Excellent product & people skills — you can translate mission requirements into testable architecture and recruit/mentor engineers.

Nice-to-have (differentiators)

• Prior work on satellite ground systems / TT&C / telemetry and mission ops.
• Experience with SAR/image processing, GDAL/Rasterio, and raster reprojection/orthorectification.
• Familiarity with ML infra: PyTorch/TensorFlow, NVIDIA CUDA stack, Triton inference server, MLflow, or KServe.
• Knowledge of spatial servers and tiling: GeoServer, pg_tileserv, Tile38, MapProxy, or similar.
• Experience with object storage (S3/MinIO), Ceph, and high-performance filesystems (Lustre).
• Security & compliance experience for govt/defence environments.

Preferred tech stack

Languages & frameworks:
Python, Go, optional)

Orchestration & CI/CD:
Kubernetes (EKS/GKE/AKS + on-prem k8s), Docker, ArgoCD / Flux, GitHub Actions / GitLab CI, Jenkins (legacy support)

Infrastructure as Code:
Terraform, Ansible, Packer

Messaging & streaming:
Kafka (Confluent), RabbitMQ (where applicable)

Databases & catalogs:
PostgreSQL + PostGIS, ClickHouse (analytics), Redis (caching), Iceberg / Delta / Parquet for analytics

Object & block storage:
S3 / MinIO, Ceph, NFS, Lustre for HPC storage

Geospatial tooling:
GDAL, Rasterio, COGs, GeoParquet, GeoServer, pg_tileserv, PostGIS, TileServer GL, Mapbox/Leaflet for frontends

Big data / distributed compute:
Spark, Dask, Ray for parallel processing; Airflow / Prefect for orchestration

ML / inference:
PyTorch/TensorFlow, NVIDIA CUDA/CUDA-aware schedulers, Triton Inference Server, MLflow / KServe

Observability & Security:
Prometheus, Grafana, ELK (ELK/Opensearch), Jaeger, Vault, Falco, OPA/Gatekeeper, SSO (OIDC)

Ground & mission ops:
Familiarity with CCSDS framing, COSMOS-style consoles, and telemetry ingestion libraries; ground station integration patterns.

Hardware & networking:
10/40/100GbE NICs, GPU servers (NVIDIA A100/V100 or similar), SAN/NAS design, UPS & rack-level considerations.

What We're Looking For

We're seeking a
builder and systems thinker
— someone who thrives at the intersection of infrastructure and intelligence. You should be equally excited about
architecting resilient DevOps systems
and
designing AI-ready platforms
capable of handling complex
geospatial and multi-sensor data flows
.

If you want to help build the foundational software layer that enables satellites and AI to work together — this is the role for you.

Why GalaxEye

• Be part of
  India's only multi-sensor satellite startup
  , building the world's first hybrid imaging constellation.
• Collaborate with top-tier engineers and scientists from IIT Madras and global space-tech communities.
• Build
  real deep-tech infrastructure
  that powers national security, global resilience, and data intelligence.