Page 39 - The 10 most Promising Artificial Intelligence Startups 2019
P. 39
“
Science / Machine learning / Deep We believe AI based LegalTech
Learning, CCNA, RHCE etc. product can bring quicker decision
Proficient in handling milieu in legal tech area and can improve
across various aspects of financial We follow Scrum Agile citizen’s life.
services and products like equity, methodology and take AI has potential to bring “Good
derivatives, fixed income bond, Health for All” in Rural Areas in
and money market product like FX regular client feedback a Developing Country like India.
Options, Rates. for product atleast once Our product can reach to detect
Our CEO: Shraddha Nagar (MBA in two weeks to improve cancer in area where doctors are
Finance Symbiosis Pune, Total unavailable.
Work experience: 10 Years) our product Deep learning and recent advances
Our COO: Saylee Begampure in computational power have
enabled the Development of
(Total Work experience: 8 Years)
She has worked with VInvent for Intelligent Medical Image
Bizamica, AIMACTEK and Biop. of plain vanilla rule-based system. Analysis Systems that can display
“
ai clients from past one year. It had helped client to onboard remarkable performance using
The Biggest Challenges We Face multiple clients and provided Artificial Neural Networks to
produced promising results when
Google cloud scalability, high
Biggest Challenges with faced availability, recovery etc inbuilt classifying skin cancer at early
by our Start-up is to identify a features. stages to enable physician to do
Business problem statement, find medical diagnosis efficiently.
target market segment, customers VInventTech Strategies to Tackle Most current methods in the field
on boarding on product developed Its Competition of skin lesion classification rely
and survive in market for long term Improvement is our strategy. on hand crafted features, such
by outperforming into existing We are improving our platform as ABCDE rule (the acronym
competition. regularly; at the same time, we stands for Asymmetry, Border,
We follow Agile Scrum are analysing what new our Color, Dermoscopic structure and
methodology with Sprint of 2 competitors are doing. We follow Evolving), and CASH (Color,
weeks to organize, plan and Scrum Agile methodology and Architecture, Symmetry, and
prioritize our work. take regular client feedback for Homogeneity). Physicians often
product atleast once in two weeks
Understanding the Clients’ Needs to improve our product. rely on personal experience and
First of all, we need to understand We also monitor if any alert raised evaluate each patient’s lesions on
a case by case basis by taking into
what our customer wants, what is in production using Circle CI and account the patient’s local lesion
their pain areas, how our expertise Sentry alert mechanism. We use patterns in comparison to that of
can help them to achieve their Test automation and nightly build the entire body.
Goals and who can we provide CI/CD methodology to improve In the past few years the number of
scalable solution to handle future our system based on feedback and
requirements of expansion. learning gather from production. skin cancer cases has been going
up and studies announce that the
Initially our customer wanted frequency of cancer doubles every
to build a rule based products Our Work on Making the 20 years. Same AI model and
Industry Better
to identify Risk associated with algorithm can be reuse to predict
Legal documents like NDA (Non- There are lots of pending Juridical other types of cancers as well e.g.
Disclosure-Agreement Contract cases in India. If we build AI based Breast Cancer.
between two parties). Lawyer who is capable of taking
We came up with solution of Deep accurate and efficient decision then Developing a Good Team
learning technology with 95% & we can solve LegalTech cases way Good People/Team Develop AI
above accuracy and build a product ahead of time. based Product. Vice versa is not
hosted on Google Cloud instead Justice delayed is Justice Denied. hold true in real life.
October 2019 39
