Founder of Irys
Founder of Irys • AI Infrastructure Expert helping organizations deploy production-ready AI that drives revenue, policy, and scale
I write high-performing code and scripts for organizations to help them generate more revenue, identify areas of investment, isolate redundancies, and automate processes.
Proven work experience in Machine Learning and Software Development. Contributions in various fields like Disease Detection, Climate Modeling, Health System Analysis, Valuable Customer Identification, DeepFake Detection, Automated Machine Learning, Supply Chain Forecasting, Text Analysis, and English to SQL translation.
Throughout my career, I have demonstrated my ability to work independently to attain results. I have been involved in projects from the data collection phase all the way to model deployment and retraining. My employers have ranged from large, multi-national corporations, state governments to small startups. This has given insight into the various intricacies of implementing effective Machine Learning and AI systems at all levels/stages. My creativity, deep technical knowledge, and effectiveness have been acknowledged by all my co-workers and can be seen through my LinkedIn recommendations.
Irys is an AI-native technology company building reasoning-driven infrastructure for legal and other high-stakes professional environments where accuracy, explainability, and trust are non-negotiable.
The company focuses on helping legal professionals, regulated enterprises, and mission-critical organizations move beyond brittle automation and generic AI tools by deploying model-agnostic reasoning systems designed for complex decision-making. Irys enables legal teams to analyze information, surface insights, and support decisions with systems that are transparent, auditable, and aligned with real-world legal workflows.
Unlike traditional AI platforms that prioritize speed over reliability, Irys is built for domains where mistakes carry legal, financial, or regulatory consequences. Its infrastructure integrates advanced machine learning with structured reasoning to support tasks such as legal analysis, risk assessment, compliance workflows, and enterprise decision support — without sacrificing accountability.
By modernizing legacy processes and embedding intelligence directly into the system layer, Irys helps legal professionals and regulated organizations reduce cognitive overload, improve decision quality, and operate with greater confidence under pressure.
Work acknowledged and used to guide Healthcare System policy for a State Government.
Automated pipelines for no-code Machine Learning solutions that raised millions in funding.
Identified high-value prospective customers with 95% accuracy, improving profits and saving resources.
Co-developed an algorithm competitive with Apple despite using fewer resources.
Content translated into multiple languages including Spanish and Malay. Students worldwide use the material.
Got 19 strangers into a tree willingly. On a Friday night. It's really hard. Try it if you don't believe me.
Open source AI community pushing the boundaries of what's possible.
Weekly insights on AI research, development, and the future of legal AI.
From fundamentals to cutting-edge applications in enterprise environments.
Building reasoning infrastructure for high-stakes legal decision-making.
The unique challenges and opportunities in building AI for the legal industry.
Deep dives into cutting-edge AI research, model development, and innovation.
You've worked across healthcare, finance, government, and startups — what's the biggest mistake organizations make when implementing AI in high-stakes environments like legal or policy decision-making?
Legal AI is often marketed as automation, but you talk about reasoning infrastructure. What's the difference, and why does it matter?
You bootstrapped to $1M ARR in just seven months. What technical or strategic decisions made that speed possible?
How do you think about building AI systems that are accurate, explainable, and trusted — especially in legal and regulated industries?
From disease detection to legal reasoning, what patterns have you noticed that separate AI projects that actually ship from those that stay stuck in experimentation?
With AI advancing so fast, what should founders and enterprises stop doing immediately — and what should they double down on instead?
How Devansh Built a Million-Strong Movement by Breaking Every Rule
DMR News coverage on AI innovation
Deep dive into AI and machine learning
Discussing machine learning operations
Featured segment at timestamp 3:57
On the future of AI in legal and beyond
FUNDING MODEL
Bootstrap
REVENUE
$1M ARR
TIMELINE
7 Months
Proprietary Technology: Model-Agnostic Reasoning Infrastructure combined with a Legal-Specific Reasoning Engine.
For media, podcast, or speaking inquiries
All requests managed through Kitcaster
© 2026 Devansh. Media kit powered by Kitcaster.