Umang Kalra

Projects

Products built for real-world complexity.

Each case shows how I bridge product strategy, technical leadership, and execution across spatial systems, operations, and customer-facing platforms.

Representative GIS workspace: estate panel and satellite map with overlays (client UI anonymized).

200+ daily active users

Enterprise supplier compliance & deforestation monitoring

Led product and technical delivery for a confidential geospatial compliance platform used by global supply-chain teams — 24M+ hectares monitored, 200+ daily active users, and materially faster reporting operations.

DjangoPostGISReactAWSMaps

DAU

200+

Reporting cycle

~60% faster

Cloud cost

~20% lower

Monitored scope

24M+ ha

Outcomes

Established one trusted operating system for compliance: 200+ daily actives, faster reporting cycles, and lower cloud spend through architecture hardening.

View project
Ranked neighborhood results with interactive map — representative PropTech UI.

58,000+ neighborhoods · sub‑second ranking

Neighborhood intelligence & preference‑driven home search

Owned product strategy and engineering for a PropTech platform ranking 58,000+ neighborhoods against user priorities in sub-second time, combining market data with location intelligence.

DjangoPostGISReactNext.jsAWS

Launch window

1,000+ users

Index size

58,000+

Sales records

900,000+

Search latency

<1s

Outcomes

Drove early traction, strong activation, and reliable performance with sub-second ranking, high uptime, and a monetization model teams could operate confidently.

View project
Map-first AOI selection with cadastral overlays — representative marketplace UI.

10,000+ users · 100+ exports/day

National geospatial data marketplace & self‑serve exports

Led a country-scale geospatial marketplace where planners and operators could discover, price, and export standardized data through one map-first workflow.

FlaskPostGISReactAWSMaps

Registered users

10,000+

Daily exports

100+

Prep time

~70% faster

Repeat usage

90%+

Outcomes

Built a durable AOI-to-delivery product flow with high repeat usage, secure transactions, and automated exports that reduced operational friction for data teams.

View project
Time‑series gas probability traces with detection thresholds — representative monitoring output (anonymized).

~90% smaller models · ~3.5× faster inference

Edge environmental & fire intelligence — ML optimization and Earth‑observation validation

Partnered with a deep-tech team to optimize an edge AI environmental intelligence stack, combining model compression, sensor stabilization, and satellite validation for field-ready reliability.

PyTorchStructured pruningINT8 / INT4NASA FIRMSGOES‑16 ABIVIIRSQGIS

Model compression

~90%

Inference speedup

~3.5×

Quantization

INT8 / INT4

EO products

VIIRS + GOES‑16

Outcomes

Delivered major model efficiency gains and more stable field behavior, with a defensible multi-source validation workflow for safety- and climate-critical operations.

View project