Hi! I’m Sri Krishna. Currently, I’m a co-founder of an AI startup working in the intersection of AI and UI-UX Design. Previously, I’ve worked as a founding engineer at ImagoAI and Paralleldots. I also collaborated with the Cognitive Vision group at Orebro Uni/Uni of Bremen. I was associated with Red Hen Lab in the Google’s SoC program for couple of years. I did my undergrad at BITS-Pilani, Goa. At Paralleldots, I mostly worked on computer vision (local features, keypoint/descriptors, few shot, fine grained, weakly supervised, object detection) applied to retail tasks. Earlier, I worked on medical imaging for the real world. Was very much interested in Multimodal ML as well before the LLM era. I listen to music a lot, and used to write about it here. Some part of this website is used to put down my thoughts and ideas into writing, so read at your own risk :)
If you want to get in touch or collaborate please reach out to me via email :).
Products
2023-Present
- Flek AI
- Built multiple products in the product design space
- Figma Plugin & NPM package to sync design and frontend code
- Diffusion based models to create product design inspirations from text prompts
- ControlNet based models to convert layout sketches into full-fledged designs based on a design system
- Parametrized model to generate visual design palletes for design systems from textual prompts
- Figma Plugin to customize screens and flows based on existing UI Kits
- Figma Plugin to generate mobile UI screens from text prompts. Using Multimodal LLMs.
- Figma Plugin to help with UI Audits using Multimodal LLMs.
- Understanding of design systems and generation of UI assets following design system guidelines.
- Product design platform with following AI Copilots
- Design System Generation, PRD Generation, Flow Generation
- Sitemap Generation, Wireframe Generation, Design Generation & Frontend Sync
2022-2023
- ImagoAI
- Founding engineer, building and deploying cutting edge hyperspectral AI models for food quality estimation.
2018-2022
- Shelfwatch
- Pivoted from Medical AI to revenue making products.
- Built the ML for the retail computer visionproduct from scratch to 1M+ ARR, alongside a brilliant team of engineers.
2016-2018
- Diagnosis IO
- Built and deployed AI models on head CT scans to help in triaging brain haemorrhage patients
- Dentistry AI
- Built and deployed AI models on bitewing X-rays to help in detecting tooth caries
Research
2023
- Visual Sensemaking Needs Both Vision and Semantics: On Logic-Based Declarative Neurosymbolism for Reasoning about Space and Motion
- Jakob Suchan, Mehul Bhatt, Srikrishna Varadarajan
- 39th International Conference on Logic Programming (ICLP 2023), 393-395
2021
- Semi-supervised Learning for Dense Object Detection in Retail Scenes
- Jaydeep Chauhan, Srikrishna Varadarajan, Muktabh Mayank Srivastava
- Paper featured at RetailVision Workshop at CVPR 2021
- [paper] [bibtex]
- Jaydeep Chauhan, Srikrishna Varadarajan, Muktabh Mayank Srivastava
2020
- Commonsense Visual Sensemaking for Autonomous Driving: On Generalised Neurosymbolic Online Abduction Integrating Vision and Semantics
- Jakob Suchan, Mehul Bhatt, Srikrishna Varadarajan
- Extended version of IJCAI paper, contributed submission at Artificial Intelligence Journal (AIJ)
- [paper] [bibtex]
- Jakob Suchan, Mehul Bhatt, Srikrishna Varadarajan
- Driven by Commonsense: On the Role of Human-Centred Visual Explainability for Autonomous Vehicles
- Jakob Suchan, Mehul Bhatt, Srikrishna Varadarajan
2019
- Benchmark for Generic Product Detection: A Low Data Baseline for Dense Object Detection
- Out of Sight But Not Out of Mind: An ASP Based Online Abduction Framework for Visual Sensemaking in Autonomous Driving
- Jakob Suchan, Mehul Bhatt, Srikrishna Varadarajan
- Accepted at International Joint Conference on Artificial Intelligence (IJCAI) 2019
- Distinguished paper honorable mention [Top 3/850 accepted papers]
- [paper] [bibtex]
- Jakob Suchan, Mehul Bhatt, Srikrishna Varadarajan
- Deep Semantics for Explainable Visuospatial Intelligence: Perspectives on Integrating Commonsense Spatial Abstractions and Low-Level Neural Features
- Mehul Bhatt, Jakob Suchan, Srikrishna Varadarajan
2018
- Multidomain Document Layout Understanding using Few Shot Object Detection
- Semantic Analysis of (Reflectional) Visual Symmetry: A Human-Centred Computational Model for Declarative Explainability
- Jakob Suchan, Mehul Bhatt, Srikrishna Varadarajan, Seyed Ali Amirshahi, Stella Yu
- Accepted in Advances in Cognitive Systems (ACS) 2018
- [paper] [bibtex]
- Jakob Suchan, Mehul Bhatt, Srikrishna Varadarajan, Seyed Ali Amirshahi, Stella Yu
- Weakly Supervised Object Localization on grocery shelves using simple FCN and Synthetic Dataset
- Srikrishna Varadarajan*, Muktabh Mayank Srivastava*
- Accepted as conference paper at Indian Conference on Computer Vision, Graphics and Image Processing (ICVGIP) 2018
- [paper] [bibtex]
- Srikrishna Varadarajan*, Muktabh Mayank Srivastava*
2017
- Anatomical labeling of brain CT scan anomalies using multi-context nearest neighbor relation networks
- Srikrishna Varadarajan, Muktabh Mayank Srivastava, Monika Grewal*, Pulkit Kumar*
- Accepted as poster at IEEE International Symposium on Biomedical Imaging (ISBI) 2018
- [paper] [bibtex] [blog]
- Srikrishna Varadarajan, Muktabh Mayank Srivastava, Monika Grewal*, Pulkit Kumar*
- RADNET: Radiologist Level Accuracy using Deep Learning for HEMORRHAGE detection in CT Scans
- Monika Grewal, Muktabh Mayank Srivastava, Srikrishna Varadarajan*, Pulkit Kumar*
- Accepted as conference paper at IEEE International Symposium on Biomedical Imaging (ISBI) 2018
- [IEEE Explore] [paper] [bibtex] [blog]
- Monika Grewal, Muktabh Mayank Srivastava, Srikrishna Varadarajan*, Pulkit Kumar*
- Detection of Tooth caries in Bitewing Radiographs using Deep Learning
- Muktabh Mayank Srivastava*, Pratyush Kumar*, Lalit Pradhan*, Srikrishna Varadarajan
- Accepted at NIPS 2017 Workshop - ML for Health.
- [paper] [bibtex] [blog]
- Muktabh Mayank Srivastava*, Pratyush Kumar*, Lalit Pradhan*, Srikrishna Varadarajan
Talks
- Weakly Supervised Object Localization on grocery shelves using simple FCN and Synthetic Dataset
- Multimodal Understanding of Videos using Machine Learning
- Presented @ Workshop on Multimodality in Embodied Experience Design
- INTERACT 2017 - The 16th IFIP TC.13 International Conference on Human-Computer Interaction, Mumbai, India
Contact
- email: skrish at protonmail dot com
- email: skrish1113 at gmail dot com
- twitter: @skrish_13
- github: @skrish13
- linkedin: @skrish13
- google scholar: link