labs

Research to Advance Intelligence.

Caelus Labs advances geospatial AI through original research. Our threads focus on SAR signal processing, multimodal reasoning, and edge perception under constraint.

[THREADS]

Research threads and active investigations

T-01 SAR Signal Intelligence→T-02 Multimodal Reasoning Systems→T-03 Edge Perception Under Constraint→

VIEW MORE →

[SIGNAL SCORE]

Live composite vulnerability index updates

1 Sahel Corridor, MaliScore: 87.4Trend: ↑

2 Ganges Delta, BangladeshScore: 83.1Trend: ↑

3 Eastern DR CongoScore: 79.8Trend: →

VIEW MORE →

[PAPERS]

Research papers and publications covering SAR intelligence, multimodal reasoning, and edge perception.

• DATE	• TITLE	• CATEGORY	• AUTHORS
6/1/2026	Any2Poster: Any-Source Poster Generation Across Modalities and Domains	Multimodal	Amogh Vinaykumar, Aiden Li, Suozhi Huang, Shilong Liu	→
3/14/2026	ALTIS: Automated Loss Triage and Impact Scoring from Sentinel-1 SAR for Property-Level Flood Damage Assessment	SAR	Amogh Vinaykumar, Prem Kamasani	→

VIEW MORE →

[BLOG]

Insights, analysis, and updates from Caelus Labs

RESEARCHJun 2026

SAR Coherence as a Proxy for Infrastructure Change

We benchmarked 18 months of Sentinel-1 coherence maps across the Sahel corridor against ground-truth infrastructure surveys.

METHODSMay 2026

Training Vision-Language Models on Overhead Imagery

A walkthrough of our data pipeline for aligning satellite patches with natural language descriptions of surface conditions.

METHODSMay 2026

Edge Inference on $400 Hardware: What We Learned

Running TensorFlow Lite models on Raspberry Pi CM4 and Hailo-8 for real-time flood boundary detection.

RESEARCHApr 2026

GRACE-FO Anomalies as Early Warning for Agricultural Stress

Mapping groundwater storage depletion three months before crop failure reports in the Indus basin.

CAELUS LABS NEWSLETTER

Research, signals, and insights, delivered to your inbox.