NJ Naman Jain
Available · Part-time
Berlin, Germany

Engineer & tech consultant.

I'm Naman Jain — a software engineer with six years of experience scaling data-intensive applications and backend infrastructure. I work at the intersection of geospatial data, machine learning and product engineering, and I'm currently taking on part-time freelance and consulting work.

Based
Berlin, Germany
Origin
Jaipur, India
Currently
Tech Lead, Marble Imaging
Domains
Geospatial · Backend · AI
My desk by the window in Berlin — laptop open to a Python project, surrounded by plants.
01

About

I thrive in bridging the gap between technical requirements and product strategy, having led teams to ship robust, high-impact features.

I grew up in Jaipur, India, studied engineering at IIT Gandhinagar, and have called Berlin home for the past five years. My background is in Python development — particularly in geospatial domains — but I work across cloud infrastructure, data pipelines and applied computer vision. Earlier in my career I built ML systems for object detection and segmentation; today I architect microservices, mentor engineers, and design clean, maintainable systems.

Outside of client work I write open-source tooling, speak at conferences like FOSS4G, and tinker on side projects — from a Deutsche Bahn fare scanner running on a Raspberry Pi to a sea-level rise simulator.

When I'm not at the laptop you'll usually find me on a bike somewhere, or with my dog Maja, or — increasingly often — working from a campervan parked in a forest. I keep an old mountain bike for the city and a steady relationship with the outdoors. None of this is on my CV; all of it informs how I work.

Working from a folding chair next to a campervan parked in a pine forest.
02

Selected work

03

Services

I take on part-time engagements where deep technical work and clear product thinking are both required.

  • i.

    Geospatial engineering

    Designing data pipelines and services around STAC, COG, Sentinel-2, GDAL/Rasterio and PostGIS. Vendor-agnostic ingestion, catalog systems, tiling, and reproducible ML workflows for aerial imagery.

  • ii.

    Backend architecture

    Python services with FastAPI on GCP / Kubernetes. Microservice design, observability with Sentry & Grafana, CI/CD with GitHub Actions and Terraform, and the unglamorous platform work that makes products durable.

  • iii.

    Applied computer vision

    Object detection, segmentation and inference pipelines with PyTorch and TensorFlow — drawing on prior work delivering CV systems at scale in production.

  • iv.

    Technical consulting

    Architectural reviews, technical strategy, hiring support and engineering mentorship — pairing with founders and product teams to bridge technical and product decisions.

04

Experience

Six years across early-stage startups and growth-stage companies.

  1. 2024 — Present Berlin

    Tech Lead

    Marble Imaging

    • Translate product requirements into technical tickets, plan sprints and assign work across the engineering team.
    • Architect microservices and drive product-led engineering initiatives connecting UX and infrastructure.
    • Design and implement core backend services handling orders, payments and automated data processing.
    • Establish internal best practices for programming, CI/CD and mentoring across the team.
  2. 2022 — 2024 Berlin

    Data Science Engineer

    UP42 GmbH

    • Developed and maintained internal Python SDKs and microservices powering customer data workflows.
    • Architected a vendor-agnostic metadata ingestion service enabling near real-time catalog updates.
    • Integrated multiple external data sources via robust adapters and standardised interfaces.
    • Key contributor to the open-source libraries up42-py and image-similarity-measures.
  3. 2021 — 2022 Jaipur (remote)

    Freelance Software Engineer

    Spacesense AI

    • Engineered Sentinel-2 data delivery pipelines and implemented Cloud Optimized GeoTIFF (COG) functionality.
    • Designed reproducible ML pipelines for aerial imagery analysis, enabling rapid experimentation and model evaluation.
  4. 2019 — 2021 Delhi

    Computer Vision Engineer

    Attentive AI

    • Built ML-powered systems for object detection and segmentation using PyTorch and TensorFlow.
    • Contributed to backend APIs and data pipelines delivering inference at scale.
    • Supported hiring and mentoring during the company's early growth phase.
05

Open source

06

Talks & writing

  • 2024 FOSS4G Europe · Tartu, Estonia

    dsm2dtm: Generate DTM from DSM for free

    Watch the talk ↗
  • 2023 125th OGC Member Meeting · Frascati, Italy

    Standardised data management with STAC

    Read the post ↗
  • 2019 FOSS4G · Bucharest, Romania

    Extracting channel heads from DEMs using machine learning

    View presentation ↗
  • Article UP42 Engineering Blog

    Bringing STAC to UP42 storage: lessons learned

    Read the article ↗
07

Skills

Geospatial
GDAL · Rasterio · PostGIS · QGIS · STAC · COG · GeoPandas · Shapely
Languages & data
Python · SQL (PostgreSQL) · FastAPI · Pandas · NumPy · PyTorch · Bash
Cloud & infrastructure
GCP · Kubernetes · Terraform · Docker · GitHub Actions · CI/CD · Sentry · Grafana
Leadership & strategy
System design · Technical strategy · Team leadership · Mentoring · Stakeholder management
08

Get in touch

For freelance enquiries, consulting work or a friendly conversation about geospatial systems — the inbox is open.