The Gen-3 AI Drug Engine
Industrializing Oncology Discovery via Fresh, Intact Human Tumor Spatial Biology
////////////////
Drug discovery is broken.
Katamaran is fixing it.
For every 20 cancer drugs in trials, 19 fail. Multi-Billion Dollar Failures.
We cure cancer in mice and petri dishes, but these results do not translate to human patients.
We are missing the tumor microenvironment (TME) and 3D architecture
Result: Al trained on this noise hallucinates better failures.
10-12
years to bring drug to the market
2x
higher success odds
70%
cost
reduction
6x
faster lead development
30x
katamaran cost efficiency for human relevant NAMs over US competitors
50,000,000+
By the numbers
50,000,000+
people worldwide are living with cancer today.
$200K vs. $60K
per year advanced drug costs:
(Developed nations vs India)
Katamaran Approach
Our Solution
Precision medicine, powered by human relevant data
Gen 1 (2013-2018)
Data: 2D cell lines, mice data, and noisy public genomic datasets
Method: Mining historical literature and flat, disconnected datasets.
Result: High quantity of targets, but poor clinical translation.
Gen 2 (2019-2023)
Data: Generative AI applied to the same noisy, public datasets.
Method: Brute-forcing structural predictions without biological context.
Result: Hallucinating novel structures on flawed, outdated biology.
Gen 3 (Katamaran)
Data: Fine tuned biological foundational models anchored by fresh, living human tumor tissue (3D tumoroids & ex vivo PCTs) (FDA NAMs) and Ex-Vivo Tumor Avatar Trials
Method: Lab-in-the-loop AI trained directly on multi-modal reality.
Result: High-fidelity, context-grounded targets, drugs, and response predictions built to survive the clinic.
*Modeled against conventional workflows; based on internal data and published benchmarks.
Our R&D and Pipeline
////////////////
-
AI-driven indication, biomarker, and responder selection.
-
Generative design for small molecules & TCEs.
-
Leveraging Bio- Foundational Models with human relevance
Predict
Tiered Approach:
-
Patient-derived tumoroids (primary screen).
-
Precision tissue slices (confirm).
-
Fine-Tuning via Proprietary Ex-Vivo Spatial Data sets (500 to 1000 Patients)
-
Ex-vivo "avatar trials" to de-risk responder hypotheses.
Validate
-
Wet lab data immediately retrains Al models.
-
Integration of patient on a chip, microfluidics in for HTS Assays,
-
Automation to scale throughput, Industrialize and shorten iteration cycles.
-
Continuous, automated retraining of Foundation Models with proprietary human data.
-
Generative De Novo Assets & Precision Repositions (High-Fidelity, High-Value IP)
Scale
(Lab-in-the-Loop)
(The Ex-Vivo Edge)
(Al-Driven)
////////////////
+
+
+
Company and Team
Pioneer in Gen-1 AI-driven oncology drug discovery since 2013( Former Founder and CEO of Lantern Pharma [NASDAQ: LTRN]). Now building the Gen-3 drug engine.
- Arun Asaithambi, Ph.D. Founder and CEO
Anand Asaithambi,
Co-Founder
Research labs
+
+
+
- Arun Asaithambi, Ph.D.
Founder and CEO
Business Model
Our Collaborators
October 10, 2017
////////////////
Featured News
Sep 26, 2023
Aug 01, 2025
October 10, 2017
For more news
October 10, 2017
What Experts Are Saying
“Head and neck cancers are extremely common but low awareness and non-specific symptoms delay diagnosis. When caught early they have excellent cure rates. Precision therapies are becoming more crucial as a one size fits all approach is ineffective.”
-Dr. Narayana Subramaniam
Lead Consultant, Head & Neck Oncology, Aster Hospitals, Bangalore
+
+
+
////////////////
We’re looking for collaborators, investors, and visionary scientists. Reach out and be part of the mission.
Quick Links
Our Address
India :
Unit, #603, 6th floor, Sigmasoft tech park, Gamma block, Whitefeild, Bengaluru, Karnataka, 560066
US :
Unit 419,180 talmadge rd, Edison nj 08817
UK :
41 Clooney Terrace, Londonderry, Northern Ireland, BT47 6AP
Mail :
Follow Us