# Rohit Lal Rohit Lal - Computer Vision / Generative AI / NLP. Current role: Computer Scientist II at NASA - ODSI (Office of Data Science) / UAH (May 2024 - Present). > Rohit Lal, Computer Scientist at NASA ODSI. Foundation models, computer vision, generative AI, and NLP. Publications at NeurIPS, ICML, ICCV, CVPR, and WACV. Website: https://rohitlal.com/ Generated: 2026-05-28. This file is built automatically from the site's data on every deploy and always reflects the current site in full. ## Summary I am currently a Computer Scientist at [NASA ODSI](https://www.earthdata.nasa.gov/esds/impact) (Office of Data Science and Informatics, formerly NASA IMPACT), where my research focuses on building and scaling AI Foundation Models for Science. I completed my MS by Research under the supervision of [Dr. Amit K. Roy Chowdhury](https://vcg.ece.ucr.edu/amit) at University of California-Riverside, where my thesis explored 3D human pose estimation for the [IARPA BRIAR](https://www.iarpa.gov/research-programs/briar) project. Prior this, I worked at IISc Bangalore's [Visual Computing Lab](https://visual-computing.in/), focusing on Unsupervised Domain Adaptation and Adversarial Vulnerability in Computer Vision algorithms. Outside of research, I enjoy traveling, photography, and table tennis. Feel free to connect! If you'd like a copy of my CV, email me take2rohit [at] gmail.com ## Highlights - Citations: 250+ - h-index: 8 - Top Venues: NeurIPS, ICML, ICCV, CVPR, WACV, IJCV - 5 publications and preprints - Recent recognition: NASA Group Achievement Award (January 2026); Research Impact Award (April 2026); Outstanding Reviewer - ICCV'25 (October 2025) ## Profiles and contact - Google Scholar: https://scholar.google.com/citations?user=q2nc3QoAAAAJ&hl=en# - GitHub: https://github.com/take2rohit/ - LinkedIn: https://in.linkedin.com/in/rohit-lal - Twitter/X: https://twitter.com/take2rohit - Email: mailto:take2rohit@gmail.com ## aboutme - Name: Rohit Lal - Area_Of_Expertise: Computer Vision | Generative AI | NLP - Github: https://github.com/take2rohit/ - LinkedIn: https://in.linkedin.com/in/rohit-lal - Email: mailto:take2rohit@gmail.com - Twitter: https://twitter.com/take2rohit - Scholar: https://scholar.google.com/citations?user=q2nc3QoAAAAJ&hl=en# - Work_Location: National Space Science and Technology Center, Huntsville, AL-35805
United States of America - Bio_Short: I am currently a Computer Scientist at [NASA ODSI](https://www.earthdata.nasa.gov/esds/impact) (Office of Data Science and Informatics, formerly NASA IMPACT), where my research focuses on building and scaling AI Foundation Models for Science. I completed my MS by Research under the supervision of [Dr. Amit K. Roy Chowdhury](https://vcg.ece.ucr.edu/amit) at University of California-Riverside, where my thesis explored 3D human pose estimation for the [IARPA BRIAR](https://www.iarpa.gov/research-programs/briar) project. Prior this, I worked at IISc Bangalore's [Visual Computing Lab](https://visual-computing.in/), focusing on Unsupervised Domain Adaptation and Adversarial Vulnerability in Computer Vision algorithms. Outside of research, I enjoy traveling, photography, and table tennis. Feel free to connect! If you'd like a copy of my CV, email me take2rohit [at] gmail.com - Stats: - - number: 250+ - label: Citations - - number: 8 - label: h-index - - number: NeurIPS, ICML, ICCV, CVPR, WACV, IJCV - label: Top Venues ## experience ### experience 1 - company: NASA - ODSI (Office of Data Science) / UAH - url: https://impact.earthdata.nasa.gov/ - title: Computer Scientist II - date: May 2024 - Present - location: Huntsville, AL, USA - description: Building and scaling AI Foundation Models for Science. - bullets: - **Weather Foundation Model:** Contributed to training [Prithvi WxC](https://research.ibm.com/blog/foundation-model-weather-climate), a 2.3B-parameter open-source weather and climate model on NASA MERRA-2 data; designed fine-tuning for autoregressive rollout forecasting, downscaling, and gravity-wave flux parameterization. [[NASA report]](https://science.nasa.gov/open-science/ai-model-weather-climate/) - **Helio Foundation Model:** Built [Surya](https://arxiv.org/abs/2508.14112), the first heliospheric foundation model on 300TB of Solar Dynamics Observatory data — explored Swin, Hiera, MaxViT, and multi-modal architectures. Close collaboration with IBM, Microsoft, and NVIDIA. - **Scaling:** Trained these foundation models across 200+ A100 GPU clusters (Jülich supercomputing, DGX) with SLURM, PyTorch FSDP, Lightning, and Hugging Face Accelerate. ### experience 2 - company: University of California, Riverside - url: https://vcg.ece.ucr.edu/ - title: Graduate Student Researcher, Visual Computing Group - date: Jan 2023 - Dec 2023 - location: Riverside, CA, USA - description: MS by Research with [Prof. Amit K. Roy-Chowdhury](https://vcg.ece.ucr.edu/amit). - bullets: - **Pose Estimation:** State-of-the-art results on occluded human pose estimation for the [STRIDE](https://sites.google.com/ucr.edu/stride/home) project, supporting downstream motion correction, prediction, and infilling. - **Biometric Recognition:** Built unsupervised pipelines for 3D/2D human pose, body-part segmentation, and silhouette extraction under occlusion for the [IARPA BRIAR](https://www.iarpa.gov/research-programs/briar) program — joint work with USC, Northwestern, and EPFL. - **Trustworthy AI:** TA for EE260B covering reinforcement learning, out-of-distribution generalization, explainability, uncertainty estimation, safety, and robustness. ### experience 3 - company: Indian Institute of Science (IISc), Bangalore - url: https://iisc.ac.in/ - title: Research Assistant, Visual Computing Lab - date: Jul 2021 - Jul 2022 - location: Bengaluru, India - description: At the [Visual Computing Lab](https://visual-computing.in/) with [Dr. Anirban Chakraborty](https://scholar.google.co.in/citations?user=NtAsZK-2HjcC&hl=en). - bullets: - **Source-free Domain Adaptation:** Proposed [CoNMix](https://sites.google.com/view/conmix-vcl), achieving state-of-the-art source-free single- and multi-target DA across standard benchmarks. - **Domain Gap Reduction:** Introduced an unsupervised [domain-gap reduction](https://link.springer.com/article/10.1007/s11263-023-01810-0) method that mitigates pseudo-label noise. *(IJCV)* - **Adversarial Vulnerability:** Developed a holistic [adversarial-vulnerability score](https://sites.google.com/view/sample-adv-trustworthy/) enabling data-efficient knowledge distillation. *(CVPR-W)* ### experience 4 - company: NUS - National University of Singapore - url: http://www.nus.edu.sg/ - title: Deep Learning Research Intern (Remote) - date: Apr 2020 - Nov 2020 - location: Singapore - description: Completed my intern under [Prof. Hongliang Ren](https://scholar.google.com/citations?user=rcF7N44AAAAJ&hl=en) at [Medical Mechatronics Lab](http://www.labren.org/mm/). My task was to do tracking of gaits generated by Origami robots. This task has been done using traditional CV techniques and the next step is to incorporate various deep learning techniques for 6D pose estimation. ## publications ### publications 1 - paper: Surya: A Foundation Model for Heliophysics - author: Sujit Roy, Johannes Schmude, **Rohit Lal**, Vishal Gaur, Marcus Freitag, Julian Kuehnert, ... , Juan Bernabe-Moreno, Rahul Ramachandran . - pub: arXiv preprint (NASA × IBM open release), 2025 - type: Preprint - paper_link: https://arxiv.org/abs/2508.14112 - project_page: https://science.data.nasa.gov/features-events/inside-surya-solar-ai-model - code_link: https://github.com/NASA-IMPACT/Surya - video: https://www.youtube.com/watch?v=KWMoF97C1Ds - abstract: A 366M-parameter open foundation model trained on 9 years of NASA's Solar Dynamics Observatory data - learns general-purpose solar representations that transfer to flare forecasting, solar-wind prediction, and active-region segmentation. ### publications 2 - paper: Prithvi WxC: Foundation Model for Weather and Climate - author: Johannes Schmude, Sujit Roy, Will Trojak, Johannes Jakubik, ..., **Rohit Lal**, ... , Manil Maskey, Tsengdar J Lee, Rahul Ramachandran - pub: arXiv preprint (NASA × IBM open release), 2024 - type: Preprint - paper_link: https://arxiv.org/abs/2409.13598 - project_page: https://research.ibm.com/blog/foundation-model-weather-climate - code_link: https://github.com/NASA-IMPACT/Prithvi-WxC - video: https://www.youtube.com/watch?v=N6d0CqTjBbo - abstract: A 2.3B-parameter atmospheric foundation model trained on MERRA-2 reanalysis that generalizes across forecasting, downscaling, gravity-wave parameterization, and hurricane-track prediction - released openly on Hugging Face. ### publications 3 - paper: Layer-wise Alignment: Examining Safety Alignment Across Image Encoder Layers in Vision Language Models - author: Saketh Bachu, Erfan Shayegani, **Rohit Lal**, Trishna Chakraborty, Arindam Dutta, ... Nael Abu-Ghazaleh, Amit K. Roy-Chowdhury - pub: ICML 2025 (Spotlight - top 2.6%) - type: Conference - paper_link: https://arxiv.org/abs/2411.04291 - abstract: Shows that harmful information in vision-language models is unevenly distributed across image-encoder layers - exposing a layer-dependent safety-alignment weakness exploitable by jailbreak attacks. ### publications 4 - paper: STRIDE: Single-Video based Temporally Continuous Occlusion-Robust 3D Pose Estimation - author: **Rohit Lal**, Saketh Bachu, Yash Garg, Arindam Dutta, Calvin-Khang Ta, D. S. Raychaudhuri, Hannah D. Cruz, M. Salman Asif, Amit K. Roy-Chowdhury - pub: WACV 2025 (Oral) - type: Conference - paper_link: https://arxiv.org/abs/2312.16221 - project_page: https://sites.google.com/ucr.edu/stride/home - code_link: https://github.com/take2rohit/STRIDE - bibtex: https://scholar.googleusercontent.com/scholar.bib?q=info:1qHf02jQ-hMJ:scholar.google.com/&output=citation&scisdr=ClF_9LdCEIet2UyO5kY:AFWwaeYAAAAAZ_6I_kZCW4yLGm_EZsKjjUP3tjs&scisig=AFWwaeYAAAAAZ_6I_jEIQlNnJmF1OhDbt01gbuM&scisf=4&ct=citation&cd=-1&hl=en - abstract: Test-time optimization that fits a per-instance optical-flow prior and an SMPL body model to a single occluded in-the-wild video - temporally smooth, occlusion-robust 3D human pose on the IARPA BRIAR benchmark. ### publications 5 - paper: VOccl3D: A Video Benchmark Dataset for 3D Human Pose and Shape Estimation under Real Occlusions - author: Yash Garg, Saketh Bachu, Arindam Dutta, **Rohit Lal**, Sarosij Bose, Calvin-Khang Ta, M. Salman Asif, Amit K. Roy-Chowdhury - pub: ICCV 2025 - type: Conference - paper_link: https://arxiv.org/abs/2508.06757 - project_page: https://yashgarg98.github.io/VOccl3D-dataset/ - abstract: 250K+ frames of realistic (not patch-overlay) occlusions with body shape, pose, segmentation, and silhouette ground truth - used to fine-tune CLIFF/BEDLAM-CLIFF for substantial gains under heavy occlusion. ## projects ### projects 1 - title: Person Follower Mobile Robot - abstract: Helper robots are widely used in various situations, for ex-ample at airports and railway stations. This paper presents a pipelineto multiplex the tracking and detection of a person in dynamic envi-ronments using a stereo camera in real-time. Recent developments inobject detection using ConvNets have led to robust person detection.These deep convolutional neural networks generally fail to run with highframes rates on devices with less computing power. Trackers are alsoused to retain the identity of the target person as well as imposefewerconstraints on hardware. A concept of multiplexed detection and tracking is used which makes the pipeline faster by many folds. TurtleBot-2 is used for prototyping the robot and tuning of the motion controller.Robot Operating System (ROS) is used to set up communication be-tween various nodes of the pipeline. The results found were comparableto current state-of-the-art person followers and can be readily usedinday to day life. - paper: https://drive.google.com/open?id=17Xxn3PumStUPc01p46W6luhm79xsPaNj - video: https://www.youtube.com/watch?v=XnrbU1050ls - github: https://github.com/khush3/person_following_bot ### projects 2 - title: Indian Number Plate Recognition - abstract: Our current solution (implemented) provides a robust registration plate detection, and extracts other features like car model, speed, face (if visible), date and time of entry/exit and upload the extracted data to a centralized IoT integrated database. Beneficiaries include malls, colleges, parking lots, etc. with multiple gates. Whenever the gate camera detects a departing car, the corresponding owner gets notified. Further, the owner can use the Alert feature to warn the guard. The web application has two levels of access, the first providing general information about a specific car to the corresponding owner, and the latter one for the Authority, which stores all the data of a campus. This can be used to monitor the traffic on the campus and for surveillance applications. - video: https://youtu.be/Y-EuzsubixI - github: https://github.com/conspicio-ai/alpr ### projects 3 - title: Solving Taxi v3 of OpenAi Gym - abstract: There are four designated locations in the grid world indicated by R(ed), G(reen), Y(ellow), and B(lue). When the episode starts, the taxi starts off at a random square and the passenger is at a random location. The taxi drives to the passenger's location, picks up the passenger, drives to the passenger's destination (another one of the four specified locations), and then drops off the passenger. Once the passenger is dropped off, the episode ends. Observations:- There are 500 discrete states since there are 25 taxi positions, 5 possible locations of the passenger (including the case when the passenger is in the taxi), and 4 destination locations. This problem was solved using the Q-Learning Approach. The model trained is consistently among the top-5 in the [OpenAI Gym Leaderboard](https://github.com/openai/gym/wiki/Leaderboard). - video: https://github.com/take2rohit/taxi_v3_openai/blob/master/result.gif - github: https://github.com/take2rohit/taxi_v3_openai ### projects 4 - title: Self balancing Camera Platform - abstract: Control system is designed to stabilize the camera gimbal system used in different airborne systems for applications such as target tracking, surveillance, aerial photography, autonomous navigation and so on. The technique is applied in everything from self-stabilizing cameras to helicopters and noise reducing equipment. This camera gimbal system replaces many traditional tracking systems such as radar which are heavy and large to mount on air vehicles. So, the stabilization of camera gimbal is very important to eliminate shakes and vibrations in photography, and provides accuracy. **NOTE:** *This project was selected for SIH-20 from our internal hackathon conducted by college. Further details will be shared after results of SIH-20* - video: https://www.youtube.com/watch?v=1D1ZQ6mKg0k - github: https://github.com/take2rohit/self-balancing-platform ### projects 5 - title: Word Embedding Generation using NLP - abstract: Word embedding is the collective name for a set of language modeling and feature learning techniques in natural language processing (NLP) where words or phrases from the vocabulary are mapped to vectors of real numbers. Conceptually it involves a mathematical embedding from a space with many dimensions per word to a continuous vector space with a much lower dimension. Word2vec is a group of related models that are used to produce word embeddings. These models are shallow, two-layer neural networks that are trained to reconstruct linguistic contexts of words. Word2vec takes as its input a large corpus of text and produces a vector space, typically of several hundred dimensions, with each unique word in the corpus being assigned a corresponding vector in the space. Word vectors are positioned in the vector space such that words that share common contexts in the corpus are located close to one another in the space. - paper: https://papers.nips.cc/paper/5021-distributed-representations-of-words-and-phrases-and-their-compositionality.pdf - github: https://github.com/take2rohit/NLP_word_embedding ### projects 6 - title: Real Time Handwritten Digit recognition - abstract: *This Summer Project was mentored by me at IvLabs, VNIT* The MNIST handwritten digit classification problem is a standard dataset used in computer vision and deep learning. Although the dataset is effectively solved, it can be used as the basis for learning and practicing how to develop, evaluate, and use convolutional deep learning neural networks for image classification from scratch. This includes how to develop a robust test harness for estimating the performance of the model, how to explore improvements to the model, and how to save the model and later load it to make predictions on new data. This was coded in scratch using Numpy. For more results and details about the algorithm visit the GitHub page. For full demo click the Video button. - video: https://www.youtube.com/watch?v=SCYhVTUIdoo - github: https://github.com/GlazeDonuts/Summer-Project-2019 ### projects 7 - title: Harmonic Motion Analyser using MATLAB - abstract: Simple harmonic motion can serve as a mathematical model for a variety of motions, such as the oscillation of a spring. With the aim of learning computer vision and MATLAB, I worked on analyzing the motion of a target-object undergoing a damped harmonic motion. The target-object was separated from the background using color thresholding and estimated as a point object. Coordinates of this point were recorded and used to estimate the parameters associated with the mathematical model of the system like maximum displacement, mean position, the velocity at different time instants. A mathematical model was estimated by fitting a curve to the recorded data using MATLAB Curve Fitting Toolbox. - github: https://github.com/take2rohit/shm-analyzer ### projects 8 - title: Hand Gesture controlled Robot - abstract: The hand gesture controlled bot is a bot which receives it commands by giving pitch and roll to hand. This is helpful for people on wheelchair who can't even move their fingers or hands.These bots are very useful in many applications like remote surveillance, military etc. Hand gesture controlled robot can be used by physically challenged people for wheelchair control .Hand gesture controlled industrial grade robotic arms can be developed. - video: https://www.youtube.com/watch?v=TX9Vsmo5vJg - github: https://github.com/take2rohit/gesturebot ### projects 9 - title: Classifier without High Level APIs - abstract: The code is written from scratch using PyTorch for data loading, matrix calculations, and GPU acceleration. This was my first introduction to DL where I wrote the code myself along with learning various mathematics and techniques required to optimize a network (PS. This also included learning ways to tune hyperparameters). Deep Learning Models Implemented are enlisted below: * Logistic Regression (Not technically a deep learning model, but gives a base for Neural Networks) * Deep Neural Network (Fully-connected layers only)


Optimization Algorithms Implemented:
* Batch Gradient Descent * Mini-batch Gradient Descent * Stochastic Gradient Descent (put batch_size = 1 in mini-batch to get this) * Batch Gradient Descent with Momentum * Batch RMSProp * Batch Adam


*
The results and algorithms are available in the GitHub repository
* - github: https://github.com/take2rohit/flower-classifier ## patents ### patents 1 - patent: Navigation System for a Vehicle and for Navigation - author: Pandya Karan, Kotecha Prakrut, Iyer V Aadhithya, Gaikwad Ravishankar, Lal Rohit, Agrawal Rishesh and Shital Chiddarwar - issued: Dec 1, 2019 - number: 201921049473 - file: https://drive.google.com/open?id=19898gw1IcNZqf98pRE90rzKPmm0TeV_G - video: https://www.youtube.com/watch?v=ePORvzIond8 ## achievements ### achievements 1 - title: NASA Group Achievement Award - description: Honored with NASA's Group Achievement Award for pioneering work on AI Foundation Models at NASA Marshall Space Flight Center. - date: January 2026 - link: https://www.linkedin.com/feed/update/urn:li:activity:7427053282780577792/ ### achievements 2 - title: Research Impact Award - description: Surya Foundation Model Team recognized with Research Impact Award at the UAH Sapphire Excellence Awards Gala. - date: April 2026 - link: https://www.uah.edu/news/news/first-uah-sapphire-excellence-awards-gala-highlights-outstanding-alumni-partners-supporters ### achievements 3 - title: Outstanding Reviewer - ICCV'25 - description: Recognized as Outstanding Reviewer at the International Conference on Computer Vision for exceptional peer review contributions. - date: October 2025 - link: https://iccv.thecvf.com/Conferences/2025/ProgramCommittee#all-outstanding-reviewer ### achievements 4 - title: ICML Spotlight Paper - description: Layer-wise Alignment paper accepted as SPOTLIGHT presentation at ICML 2025, ranking in the top 2.6% of all submissions. - date: July 2025 - link: https://www.linkedin.com/feed/update/urn:li:activity:7323783710875488257 ### achievements 5 - title: Smart India Hackathon Winner - description: First position at Smart India Hackathon 2020 conducted by MHRD, winning cash prize of ₹1,00,000. - date: August 2020 ## news - k: 10 - news: - - date: Apr'26 - description: Our Surya Foundation Model Team has received the [Research Impact Award](https://www.uah.edu/news/news/first-uah-sapphire-excellence-awards-gala-highlights-outstanding-alumni-partners-supporters) - - date: Mar'26 - description: Delivered a lightning talk on an AI foundation model and was on the Deep Partnership (with NVIDIA) panel at the [NAIRR Pilot'26](https://events.internet2.edu/website/89730/home) Annual Meeting. - - date: Jan'26 - description: Our AI Foundation Model team was honored with NASA's Group Achievement Award at an award ceremony held at NASA Marshall Space Flight Center. [LinkedIn Post](https://www.linkedin.com/posts/rramachandran05_nasa-aifoundationmodels-openscience-activity-7423369334095400960-ONQ2?utm_source=share&utm_medium=member_desktop&rcm=ACoAACTSE2EBtu9uiggBxIziWSeskj6z4VMn6Z0) - - date: Oct'25 - description: Recognized as an Outstanding Reviewer at [ICCV'25](https://iccv.thecvf.com/Conferences/2025/ProgramCommittee#all-outstanding-reviewer) - - date: Aug'25 - description: Released [Surya](https://github.com/NASA-IMPACT/Surya) - Foundation model for HelioPhysics [[NASA](https://science.data.nasa.gov/features-events/inside-surya-solar-ai-model)] [[IBM](https://www.youtube.com/watch?v=KWMoF97C1Ds)] [[ArXiv](https://arxiv.org/abs/2508.14112)]. - - date: Jul'25 - description: One paper [Layer-wise Alignment](https://arxiv.org/abs/2411.04291) has been accepted as a SPOTLIGHT at [ICML 2025](https://icml.cc/Conferences/2025) (Top 2.6%). - - date: Jun'25 - description: One paper VOccl3D: A Video Benchmark Dataset for 3D Human Pose and Shape Estimation under real Occlusions has been accepted at [ICCV 2025](https://iccv.thecvf.com/Conferences/2025). - - date: Apr'25 - description: Serving as a reviewer for [CVPR 2025](https://cvpr.thecvf.com/Conferences/2025), [ICLR 2025](https://iclr.cc/Conferences/2025) and [ICCV 2025](https://iccv.thecvf.com/Conferences/2025). - - date: Feb'25 - description: Delivered an oral presentation of my Master’s thesis [STRIDE: Single-video based Temporally Continuous Occlusion-Robust 3D Pose Estimations](https://sites.google.com/ucr.edu/stride/home) at [WACV 2025](https://sites.google.com/ucr.edu/stride/home). - - date: Sep'24 - description: Our team released Prithvi WxC, a new general-purpose AI model for weather and climate. [[NASA Article]](https://science.nasa.gov/open-science/ai-model-weather-climate/) [[IBM Blog]](https://research.ibm.com/blog/foundation-model-weather-climate) [[HuggingFace CEO interview]](https://www.youtube.com/watch?v=N6d0CqTjBbo&t=28s) - - date: May'24 - description: Joined [NASA ODSI](https://www.earthdata.nasa.gov/esds/impact) as a Computer Scientist II. - - date: Jan'24 - description: Started working as a Teaching Assistant for a course on [Trustworthy AI](https://sites.google.com/ucr.edu/ee260b-w24/home) by [Prof. Jiachen Li](https://jiachenli94.github.io/). - - date: Nov'23 - description: One paper on Self-Supervised Human Silhouette Extraction under Occlusion got accepted at WACV-2024 conference [[LinkedIn Post]](https://www.linkedin.com/posts/rohit-lal_ai-machinelearning-computervision-activity-7150061004364615680-M9pw?utm_source=share&utm_medium=member_desktop). - - date: July'23 - description: One paper on Source free Pose Adaptation got accepted at [ICCV-2023](https://iccv2023.thecvf.com/) conference! - - date: Mar'23 - description: Our work at IISc on *Improving Domain Adaptation through Class Aware Frequency Transformation* got accepted to the [International Journal of Computer Vision (IJCV)](https://www.springer.com/journal/11263). - - date: Oct'22 - description: Our paper on *CoNMix for Source-free Single and Multi-target Domain Adaptation* got accepted at the [Winter Conference on Applications of Computer Vision (WACV) 2023](https://wacv2023.thecvf.com/home). - - date: Apr'22 - description: Paper on [Holistic Approach to Measure Sample-level Adversarial Vulnerability and its Utility in Building Trustworthy Systems](https://sites.google.com/view/sample-adv-trustworthy/) got accepted at [HCIS Workshop, CVPR'22](https://sites.google.com/nycu.edu.tw/hcis/home). - - date: Oct'21 - description: Our paper [Open-Set Multi-Source Multi-Target Domain Adaptation](https://ivlabs.github.io/os-nsmt/) got accepted at [Pre-registration Workshop, NeurIPS'21](https://preregister.science/). - - date: Aug'21 - description: Our work at IISc on [CAFT- Class Aware Frequency Transform for Reducing Domain Gap](https://vclab-dev.github.io/CAFT/) got accepted at [TradiCV Workshop, ICCV'21](https://sites.google.com/view/tradicv). - - date: Jul'21 - description: Joining as a Project Assistant at [Visual Computing Lab, IISc Bangalore](http://visual-computing.in/home/). - - date: Nov'20 - description: Completed my intern work at [NUS - National University of Singapore](http://www.nus.edu.sg/) on 6DOF origami pose estimation. Check out the [Video Here](https://youtu.be/59ahQ5n6Snk) - - date: Aug'20 - description: First position holder of [Smart India Hackathon 2020](https://www.sih.gov.in/) conducted by MHRD winning cash prize of Rs. 1,00,000 [[Video Link]](https://youtu.be/Y-EuzsubixI). - - date: Aug'20 - description: Collaborated with a senior scientist at [CAIR Lab, DRDO](https://drdo.gov.in/labs-establishment/about-us/centre-artificial-intelligence-robotics-cair) on drone image tracking, detection and pose estimation. - - date: Jun'20 - description: Made two GitHub repository [IvLabs/resources](https://github.com/IvLabs/resources) and [IvLabs/ResearchPaperNotes](https://github.com/IvLabs/ResearchPaperNotes) which started to gain popularity! - - date: Feb'20 - description: Bagged two first position in two different competitions named 'Techno.Docx' and 'Electroblitz' at [AXIS'20, VNIT | Central India's Largest Techfest](https://www.instagram.com/axis_vnit/?hl=en). ## skills - categories: - - title: ML & AI Frameworks - skills: - Python - PyTorch - TensorFlow - HuggingFace - OpenCV - scikit-learn - Pandas - - title: Distributed Training - skills: - FSDP - DDP - ZeRO - Pipeline Parallel - LoRA - RLHF - FlashAttention - Megatron-LM - - title: HPC & Infrastructure - skills: - SLURM - PBS - NASA NAS - Nvidia DGX - Jülich SC - AWS - Docker - - title: Development Tools - skills: - Git - Wandb - Matplotlib - Jupyter - VS Code - Linux - LaTeX - MATLAB ## courses ### courses 1 - title: Machine Learning by Andrew Ng - certificate: https://www.coursera.org/account/accomplishments/verify/G9XX9CYSFJUY - source: Coursera - source_url: https://www.coursera.org ### courses 2 - title: Neural Networks and Deep Learning - certificate: https://www.coursera.org/account/accomplishments/verify/G9XX9CYSFJUY - source: deeplearning.ai - source_url: https://www.coursera.org ### courses 3 - title: Improving Deep Neural Networks:- Hyperparameter tuning, Regularization, etc - certificate: https://www.coursera.org/account/accomplishments/verify/K5CL3SVXLY7Q - source: deeplearning.ai - source_url: https://www.deeplearning.ai/ ### courses 4 - title: Convolutional Neural Networks - certificate: https://www.coursera.org/account/accomplishments/certificate/7ADJEDF2ETRH - source: deeplearning.ai - source_url: https://www.deeplearning.ai/ ### courses 5 - title: Structuring Machine Learning Projects - certificate: https://www.coursera.org/account/accomplishments/verify/YWUNM3JRUZJW - source: deeplearning.ai - source_url: https://www.deeplearning.ai/ ### courses 6 - title: Introduction to Computer Vision - source: Aaron Bobbick | Udacity ### courses 7 - title: Sequence Models - source: deeplearning.ai - source_url: https://www.deeplearning.ai/ - instructor: Andrew Ng - instructor_url: https://www.coursera.org/instructor/andrewng ### courses 8 - title: Multivariate Calculus, Integral and Discrete Transforms ### courses 9 - title: Statistical Analysis, Probability and Queuing Theory ## travel - cities: - - name: Huntsville - description: NASA ODSI - Research Scientist - type: city - home: true - - name: Riverside - description: UC Riverside - Master's Degree - type: school - - name: New York - description: Business & Culture Hub - type: city - - name: Los Angeles - description: Tech & Entertainment - type: city - - name: San Diego - description: Pacific beaches and the country's best zoo - type: city - - name: Chicago - description: Midwest Adventures - type: city - - name: Seattle - description: Pacific Northwest - type: city - - name: Boston - description: Academic Excellence - type: city - - name: Atlanta - description: Southern Charm - type: city - - name: San Francisco - description: Innovation Valley - type: city - - name: Nashville - description: Music City Vibes - type: city - - name: New Orleans - description: Culture & Jazz - type: city - - name: Washington DC - description: Nation's Capital - type: city - - name: Orlando - description: Sunshine State - type: city - - name: Las Vegas - description: Desert Lights - type: city - - name: Death Valley - description: Hottest place on Earth, salt flats and silence - type: park - - name: Mount Rainier - description: Glacier-capped volcano looming over the Cascades - type: park - - name: Zion - description: Red-rock cathedrals you can hike right through - type: park - - name: Bryce Canyon - description: An amphitheatre of orange hoodoos at sunrise - type: park - - name: Grand Canyon - description: A mile-deep crack painted layer by layer - type: park - - name: North Cascades - description: American Alps with almost no crowds - type: park - - name: Great Smoky Mountains - description: Misty blue ridges and roughly 1,500 black bears - type: park - - name: Saguaro - description: Forests of 40-foot cacti that wave back - type: park - - name: Joshua Tree - description: Twisted yuccas and house-sized boulders at dawn - type: park