Website:
carnot.co.in
Job details:
Company: Carnot Technologies (Mahindra Group)
Location: Marol, Andheri, Mumbai · Work from office
About Carnot
Carnot was founded in 2015 with one ambition: build world-class products
out of India. Eleven years on, we're a Mahindra-backed agri-tech company
processing 20M+ pings daily on our telematics backend. With telematics
established as our core, we're now building several products on top —
telemetry-driven data analytics, satellite intelligence for agriculture,
solutions for sugar mills, dealer networks, and more. We build for
farmers, mills, fleet operators, and agri-businesses across India.
The role
For our Smart Harvesting Solution (SHS), we turn satellite imagery, weather
feeds, and parcel-level farm data into predictions for sugar yield, harvest
detection, and crop health. The data is messy on purpose: farms without
perfect boundaries, weather that refuses to behave, satellite images 80%
cloud-covered exactly when you need clear data. If that sounds exciting
rather than scary, you'll feel at home here.
What you'll build
You'll own one workflow end-to-end — data → experiments → models → APIs →
production pipelines — and own that path for at least one workflow.
Concretely:
- Build and maintain pipelines for satellite imagery, weather data, and
parcel-level farm information
- Run Python experiments to test hypotheses around sugar yield, harvest
detection, and crop monitoring
- Convert useful experiments into production pipelines, APIs, or scheduled
jobs — the kind that don't break at 2 AM
- Help us build modular components that serve SHS today and the broader
Satellite Intelligence Platform tomorrow.
This is neither a notebook-only role nor a pure ML research role. We want
someone who is genuinely happy doing both halves.
You should apply if you have
- 3+ years of professional Python experience in a production environment
(not internships, not coursework — actual systems running for real users)
- Hands-on comfort with Pandas, NumPy, and wrangling messy real-world data
- Working knowledge of at least one cloud platform: AWS / GCP / Azure
- Familiarity with PostgreSQL/PostGIS, S3/GCS, or similar data systems
- At least one concrete example, recent, of an experiment or service you
personally took from idea to production
- The ability to learn fast and run with limited hand-holding — we're a
small team and we expect you to take ownership from week one
Nice to have
- Experience with satellite imagery, remote sensing, GeoPandas, rasterio,
GDAL, or any geospatial work
- Experience with weather, time-series, or agronomic data
- Comfort using AI coding tools (Claude, Cursor, Codex, etc.) as part of
your daily workflow — we use these heavily, primarily Claude
Culture
We're an 8-person engineering team with no hard sub-teams. We love
whiteboards — the ability to explain something from first principles is
one of the highest signals we look for, in candidates and in teammates.
We focus on shipping over staying in research phase.
Click on Apply to know more.