×
0 followers0 following
Hrithik Sanyal
California, United States
Product Designer who turns complex technical systems into AI-powered experiences that scale.0 followers0 following
- About
- High School
- Bachelor's
- Master's
- Target
- TOEFL
- GRE
Male
California, United States
Work ExperienceProduct Designer
Catastrophic Wildfire Prevention Consortium (CWPC) - Full Time
Jul 15, 2024 - Dec 15, 2025 · 1 yr 5 mos
California, United States
Led UX design and research for CWPC platform including donate feature system, community scorecard, sponsorship packages, authentication flows, and design system with 200+ WCAG-compliant components. Collaborated with fire officials, community leaders, engineering, marketing, and executive teams through stakeholder workshops and usability testing to design Scorecard and donation platform through A/B testing, achieving 85% task success rate, 32% conversion increase, 25% recurring contributions, and improved onboarding completion from 60% to 82%
UX Researcher & Designer
Bound International - Full Time
Jan 9, 2023 - May 1, 2023 · 4 mos
Michigan, United States
Led end-to-end UX research and redesign for study abroad platform serving 10,000+ prospective students annually, conducting stakeholder interviews, surveys, heuristic evaluation, and competitive analysis of 6 platforms to identify navigation barriers and content discoverability issues. Collaborating with Bound International product team and conducting usability testing with 5 study abroad students plus 15 additional remote participants, diagnosed 5 critical usability issues through Nielsen's heuristic principles and developed redesigned information architecture with streamlined navigation, search functionality, and mobile-responsive layouts. Delivered comprehensive redesign specifications and interactive prototypes making key program information 40% faster to locate, with recommendations adopted for Q2 2023 implementation addressing mobile usability issues affecting 65% of traffic.
User Experience Researcher
NCID: Spark Publications (University of Michigan College of Literature, Science, and the Arts) - Full Time
Aug 29, 2022 - Dec 12, 2022 · 4 mos
Michigan, United States
Led comprehensive UX research initiative for NCID: Spark Publications editorial platform, conducting contextual inquiry across 50+ hours of workflow observation to identify automation opportunities and collaboration inefficiencies. Collaborating with editors, content strategists, and operations teams, designed and facilitated 15 semi-structured interviews, utilized affinity diagramming to synthesize 250+ qualitative insights, and mapped complete editorial workflows revealing that redundant manual tasks consumed 40% of project time and collaboration tool gaps caused 15-20% publishing delays. Developed journey maps, wireframes, and interactive prototypes demonstrating AI-assisted content tagging and centralized editorial dashboard with real-time versioning, presenting comprehensive research report with implementation roadmap to executive stakeholders, with Phase 1 recommendations (improved search and automated approvals) prioritized for 2023 Q1 development reducing projected editorial bottlenecks by 35%.
Electronics Systems Intern
Bharat Electronics Limited - Internship
Jul 3, 2017 - Nov 3, 2017 · 4 mos
Maharashtra, India
Worked as electronics engineering intern across production, quality assurance, and procurement departments at India’s leading defense electronics manufacturer. Collaborating with senior engineers and production teams, rotated through multiple divisions to understand end-to-end manufacturing workflows including railway systems, platform sliding doors, and cable assembly operations. Learned technical design tools including ActCAD and SolidWorks for component design, participated in site visits to observe production processes, and gained understanding of procurement cycles, tendering procedures, and quality assurance protocols for defense-grade electronics systems.
Product Designer
Kryptonas Innovations - Full Time
Feb 3, 2020 - Jun 14, 2022 · 2 yrs 5 mos
Delhi, India
Responsible for 0-to-1 design of e-pharmacy platform including onboarding, prescription management, checkout, MedScope design system with 60+ components, chronic care management, payment optimization, and pharmacy dashboards. Collaborating with co-founders, engineering, pharmacy partners, and medical advisors, grew platform from 0 to 10,000+ users with 68% retention, reducing prescription upload drop-off by 40%, and achieving 45% LTV increase.
Product Designer
General Motors - Full Time
Jan 8, 2024 - May 6, 2024 · 4 mos
Michigan, United States
Designed AI-enhanced luxury infotainment interfaces for high-performance driving balancing AI capabilities with NHTSA compliance. Collaborated with engineering, product managers, and safety specialists through design sprints and usability testing with 50+ participants across 3 iterations, achieving 85% task success rate and 30% reduction in user interactions while maintaining compliance standards.
Graduate Student Instructor (GSI) for SI 364 - Building Data-Based Applications
University of Michigan - Full Time
Aug 30, 2023 - May 1, 2024 · 8 mos
Michigan, United States
Taught foundational web development to 238 students covering Python, Django, HTML, CSS, SQL, and API integrations through structured lab sessions, office hours, and individualized debugging support. Collaborating with lead instructor and teaching team, guided students through full-stack application development using MVC pattern and ORM, explaining RESTful API design, data normalization, and authentication principles through hands-on problem-solving.
Helped 90% of students successfully deploy Django applications with proper database models and authentication, with class achieving 15% higher average project scores compared to previous semester through structured debugging methodology and detailed code review feedback.
Graduate Student Instructor (GSI) for SI 658 - Information Architecture
University of Michigan - Full Time
Jan 10, 2023 - Apr 18, 2023 · 3 mos
Michigan, United States
Facilitated information architecture workshops for graduate students, guiding analysis of real-world objects to understand IA principles including thingness, ontological granularity, and worldhood through weekly modeling exercises. Collaborating with lead instructor, broke down complex theoretical readings from Heidegger, Bogost, and Garrett into applied concepts, led small-group discussions, and provided individualized feedback on relational models helping students connect abstract IA theory to practical application. Students achieved 85% improvement in modeling complexity across the semester, with final projects demonstrating sophisticated understanding of context-dependent meaning and structural thinking applicable to digital information systems.
Information Architecture Intern
The Understanding Group - Internship
Jul 5, 2023 - Dec 18, 2023 · 6 mos
Michigan, United States
Conducted IA research across 15+ clients in healthcare, retail, and financial services, creating wireframes, site maps, and content taxonomies. Led card sorting and tree testing with 150+ participants, delivering 8 strategic deliverables that shaped client digital experiences.
ProjectsHrithik Sanyal has not added any project yet.
Research PapersVoice Assisted Home Automation System
Mar 18, 2018 - International Journal of Industrial Electronics and Electrical Engineering (IJIEEE)
Automation Systems mainly focuses on individuals having physical difficulties. The proposed Voice Assisted Home Automation System empowers individuals to manage electronic equipment’s and to control gadgets using individual's tone/voice. Home Automation deals with interfacing the gadgets such as lights, TVs, cooling systems etc. with the Linux operating system based primary device(raspberry pi), by means of transfer switches, the framework reacts to the individual's tone changes and cleverly activate, inactivate and modulate the condition of the appliances. The research work explains the genuine usage of the minimal effort gadget, it also demonstrates the outcomes of the framework.
VM Load Balancing Techniques for Cloud
Jul 14, 2019 - International Journal of Modern Engineering and Research Technology (IJMERT)
Cloud computing has spread over the Internet very deep and overriding technique among all parallel technologies in the current era. Major reasons for the same are available and flexible services being received from it. It is all due to the use of virtual machines for managing the user-specific services for cloud operations. Although SaaS, PaaS, IaaS have provided a great service model for the cloud, there are many challenges too. The most important challenge occurs due to availability of services using virtual machines. For virtual machines to work smoothly a lot of migration processes occurs due to variable loads which may increase or decrease over the cloud. When load decreases it all goes smoother but when load increases on a particular virtual machine or more than one virtual machines then a load balancer should be applied which will carry the responsibility of choosing a specific virtual machine and migrate it on a free server. It means there are three major challenges i.e. proactively identifying the load enhancement, selecting a specific virtual machine and then migration of the same to another server. In this paper, these problems are being discussed thoroughly and a better solution is being provided for the same. The work shall provide the high performance and accuracy of the load balancing of the virtual machines over the cloud.
Email Spam Detection Using Text Classification
Dec 2, 2019 - International Journal of Modern Engineering & Management Research (IJMEMR)
Spam has been a major problem in any online system and is available in different formats. Majorly spam creates a lot of problems in the email system. Since spam is unexpected and unwanted and is sent by the spammers to promote, hack or send malicious contents to the recipients, therefore, they create major problems such as wastage of network resources, wastage of time, damage of PC’s and laptops due to viruses security breach, mail quota problems, irritations to the recipients, ethical issues etc. Identifying and filtering SPAM mails from all the emails of the users has become a real problem as SPAM mails cause several problems for the users. SPAM mails recognition is a problem of data mining as a user may receive hundreds of emails in a day and a few of them are SPAM mails. This work is focused on clustering of Enron Data set for SPAM and NON-SPAM cluster formation from small data set to large data sets. This work finds the possibilities of providing high performance to real-time email service providers.
Application of Rules and Authorization Key for Secured Online Training—A Survey
Mar 31, 2021 - Proceedings of International Conference on Sustainable Expert Systems
Online training has been present for over a decade, and its importance is increasing every day. Today, it has become very important due to ongoing COVID-19 pandemic. As it has been accepted widely by the educational systems; nowadays, challenges like hardware resources, network resources, software resource, and security have become more demanding. Security threats among all challenges require more researches to develop rigid systems where data of all stakeholders remain secured. ML has a proven track record to solve such problems. In terms of security, ML continuously learns by analyzing data to find patterns so unauthorized access to encrypted traffic is detected better and find insider threats to keep information safe. Here, a new system is being developed using an improved algorithm, described in proposed work. Using this new algorithm, machines are trained to identify unauthorized access attempts and stop them from stealing data even if, they are authenticated.
Enhanced Accuracy in Machine Learning Using Feature Set Bifurcation
Mar 31, 2021 - Cybernetics, Cognition and Machine Learning Applications
In the diagnosis of deadly diseases, the use of machine learning is proving to be very much helpful and provides diagnosis with high performance. It has been observed that most of the deadly diseases have similar symptoms in patients. Machine learning advantage is being used in comparing the data of patients with similar symptoms, which eases decision making. Since the parameters related to the symptoms are prodigious and historical data of the millions of the patients cannot be compared with conventional processing techniques; hence, research is continuously being done to evolve techniques and algorithms for the same. This work provides an analysis of the machine learning techniques available and their applications (Sanyal and Agrawal in Int Res Anal J 14:348–353, 2018, [1]). In this work, a bifurcated feature set has been used to implement decision trees using a multithreaded environment. Paper elaborates the technique applied for bifurcation of dataset features, multithreaded decision trees implementation and combining process to obtain a final result. Implementation results depict that large feature sets are having considerable improvements in obtaining better accuracies. This improvement in the algorithm is not only enhancing the accuracy of disease prediction, but also performance.
Innovative Approach for Prediction of Cancer Disease by Improving Conventional Machine Learning Classifier
Mar 31, 2021 - Cybernetics, Cognition and Machine Learning Applications
An application of computers in medical sciences is becoming boon for the patients. Specifically, in the diagnosis of deadly diseases, it has asserted to be very much helpful and provides diagnosis with high performance. It has been observed that most of the deadly diseases have similar symptoms in patients affected by them. Comparing the data of patients with similar symptoms eases decision making. Since the parameters related to the symptoms are prodigious therefore, the historical data of the millions of the patients cannot be compared with conventional processing techniques. Hence, research is continuously being done to create new techniques and algorithms for the same. This work is a continuation of author’s work which provides an analysis of the machine learning techniques available and their applications (Int Res Anal J 14(1):348–353, [1]). Work carries the research by involving innovation to implement a modified decision tree classifier algorithm. The improvement in classifiers-based ML will not only enhance the performance of disease prediction but will also be more and more accurate due to its adaptive characteristic.
Study of Holoportation: Using Network Errors for Improving Accuracy and Efficiency
Mar 31, 2021 - Proceedings of International Conference on Sustainable Expert Systems
Holoportation is a technique in which two persons communicate with each with anyone’s virtual presence in front of the other. There have been efforts to make it as much smoother and real as possible by the researchers in the recent past. But the challenges are many in this field not only due to unavailability of software resources but due to hardware constraints as well. Major hardware constraints are based on the transmission of a lot of data being collected by the camera and audio devices which require good data transfer rates between the communicating devices. Reason of challenge is viewed in two faces, i.e., one is slow data transfer speed and the second is huge amount of data transfer. Slow data transfer speed of resources is being tackled, and a good data transfer rate has been reached to but still not suffice and unavailable in all areas around the world. A huge amount of data transfer may also suffer from network lag spikes and dropouts of the signals which will lead to disruption in reproduced Holoportation on the receiver’s end. In this paper, the focus is on proposing a buffering and correction mechanism which will require to be applied on both sender and receiver ends. The system will produce high accuracy and will not increase network latency and hence the smoothness of service. The system will leverage normal human behaviour and persistence of vision delays to provide better accuracies.
Study of Bot Technology to Provide High Performance and Accuracy using Slam Technology
Nov 10, 2020 - 2020 Fourth International Conference on I-SMAC (IoT in Social, Mobile, Analytics and Cloud) (I-SMAC)
As the resources over the internet are increasing day by day in terms of both hardware and software, information stored over them is also increasing exponentially. Related information stored over the Internet is not only increasing but it is widespread around the world on different resources. Search engines e.g. Google, Bing, Yahoo etc. are listing this information to the users based on their own designed algorithms which require ways to associate information providers to these search engines services. This mechanism has two major drawbacks, first, it bounds to the information providers to provide permissions forcefully to the search engine service provider and second is they are generalized and users get lost in huge information retrieved by the search services. Along with this, the information retrieval is time-consuming as well. This work is focusing to provide high performance, high accuracy, customization-based search tool which leverages the advantages of multiple technologies.
Improved Rules and Authorization Key Processing for Secured Online Training
Dec 28, 2020 - 2020 4th International Conference on Electronics, Communication and Aerospace Technology (ICECA)
The security aspect is considered as most important for any system especially those systems which include user’s data. It is having multiple mechanisms to be applied primarily authentication and authorization. This requires applying better mechanisms for authentication such as encryption, OTP, etc. whereas for authorization RBAC and other protocols have been established and used. Researchers have found that there is a lot of scope for improving the RBAC and other protocols along with developing customized protocols for enhanced security using authorization. In the previous work to this paper a new enhanced authorization mechanism has been proposed which uses a customized set of rules and authorization key-based mechanisms to improve security and have got a lot of appreciations. In this work, implementation of this mechanism is being provided with results obtained which are found to be better than the existing systems.
Use of SLAM Technology to Enhance Performance & Accuracy of Bots
Mar 1, 2021 - 2020 2nd International Conference on Advances in Computing, Communication Control and Networking (ICACCCN)
Bot technology is proving to be a boon in the current era when Internet has become a necessity for the world. With the advantage that it collects remote data with authentication and authorization using the various automation features, it has added not only to collect data but also to provide it to the desiring users with lot of ease. As now a days the data usage and storage is increasing continuously there is a need for a more accurate and appropriate method which can provide customized and relevant data with respect to the specific need, current algorithms used by search engines google, yahoo and Bing is providing the data in an un-customized manner and it is also very time consuming , due to which the user get lost in the huge data and has to deal with time latency as well. To overcome this, we have proposed a general solution in this paper which can provide better filtering of data specific to the information needed, reduce time latency by using various technologies and algorithms with better accuracy.
Natural Language Processing Technique for Generation of SQL Queries Dynamically
May 10, 2021 - 2021 6th International Conference for Convergence in Technology (I2CT)
Natural Language Processing is being used in every field of human to machine interaction. Database queries although have a confined set of instructions, but still found to be complex and dedicated human resources are required to write, test, optimize and execute structured query language statements. This makes it difficult, time-consuming and many a time inaccurate too. Such difficulties can be overcome if the queries are formed dynamically with standard procedures. In this work, parsing, lexical analysis, synonym detection and formation processes of the natural language processing are being proposed to be used for dynamically generating SQL queries and optimization of them for fast processing with high accuracy. NLP parsing of the user inputted text for retrieving, creation and insertion of data are being proposed to be created dynamically from English text inputs. This will help users of the system to generate reports from the data as per the requirement without the complexities of SQL. The proposed system will not only generate queries dynamically but will also provide high accuracy and performance.
Study of Depression Detection using Deep Learning
May 13, 2021 - 2021 IEEE International Conference on Consumer Electronics (ICCE)
Deep learning is becoming an important method of machine learning and has proven to provide its strength in data sciences and artificial intelligence. It works with unstructured and unlabeled data. Deep learning is more accurate than conventional classifier-based machine learning. The reason for high accuracy is managing by a count of hidden layers, application of activation functions, weighted activation functions, etc. In this work, a new weighted activation function is being proposed to get high accuracy for deep learning. The weighted activation function is going to be applied and tested on a depression dataset to evaluate the intensity of depression.
VolunteeringHrithik Sanyal has not added any volunteering detail yet.