Help & Supportcareers@gravityer.com
AI @ Uniphore | GaTech CS | IIT-Kanpur CS | ex-{MSFT, Google, Orby} AI
AI is analyzing your overall score…
Identifying your key strengths…
Evaluating your skill match against the job requirements…
Assessing your cultural and operational fit
I am passionate about developing disruptive products that have the potential to make a direct impact in the industry. I believe in smart and neat execution, starting from profound research in subject matter to robust implementation.
Georgia Institute of Technology
Master's degree, Computer Science
January 1, 2015 – January 1, 2017
Indian Institute of Technology, Kanpur
Bachelor of Technology (B.Tech.), Computer Science
January 1, 2011 – January 1, 2015
Uniphore
Senior Staff Software Engineer
September 1, 2025 – Present
Greater Seattle Area · Remote
Orby AI
Software Engineer
November 1, 2024 – Present
Greater Seattle Area · Remote
Senior Software Engineer (ML)
October 1, 2023 – October 1, 2024
On-site
Software Engineer
March 1, 2019 – November 1, 2023
On-site
Microsoft
Software Engineer
July 1, 2017 – March 1, 2019
Redmond, USA · On-site
Apple
Software Engineer
May 1, 2016 – July 1, 2016
Cupertino, US
Microsoft India
Software Developer
May 1, 2014 – July 1, 2014
Hyderabad Area, India
ArtZolo
Developer
December 1, 2012 – February 1, 2013
Estimating Bias of Media towards Presidential Candidates on Twitter
March 1, 2016 – Present
Used IBM Watson Alchemy Language API for Sentiment Analysis of collected tweets and assigned a bias score to ten prominent media houses towards the candidates in the current US presidential race.
Gaze Detection
March 1, 2016 – Present
Given an image containing one or more person, predict the location where each person is looking at. Image is passed through two parallel deep learning pipelines with one detecting salient objects and other the direction of gaze. Then combine both to detect the object person is looking at in the image. Also, experimented with geometry based techniques to fetch the gaze direction.
Loop Detection in Video
February 1, 2016 – Present
Detect previously visited places during travel of the mobile robot. Passed two images in the sequence through a siamese network. Siamese network consist of pre-trained Places-CNN and Alexnet network fine tuned on specific examples. Used Contrastive Loss as the loss function as it encourages matching pairs to be close together in feature space while pushing non-matching pairs apart.
Hierarchical Web Classification
September 1, 2014 – April 1, 2015
Developed a hierarchical classification by extending the FastXML: a multi-label classifier and iterating it over a static hierarchical structure of DMOZ dataset. - Integrated LDA Topic Modelling to discover the latent topics. - Studied the correlation between f-measure and clusters formed at each level; compared various techniques such as Clustering models, RFM models, SVM models for classifying web pages in a hierarchy structure.
Peer-to-Peer Concurrent Editing using Chord DHT
September 1, 2014 – November 1, 2014
Developed a peer-to-peer concurrent rich text editing platform using Chord DHT(Distributed Hash Table). Maintained 2 layer overlay network of chordal ring and k-ary tree to broadcast each message in O(logn) time. Extended Distributed Operational Transform algorithm to ensure the consistency of copies being concurrently updated in distributed nodes.
Humanoid Robot: Throw Ball at a Target
January 1, 2014 – April 1, 2014
Project aims to identify known target through ALDEBARAN NAO and then to throw an object towards it. we first detect the known target using Image Processing techniques, then we compute the position of that object and finally we perform the throwing action. - Developed end-to-end NAOqi modules for detecting known target in cluttered environment and joint control approach to throw ball towards it. - Implemented learning based Linear Approximation Model to estimate the parameters correctly and Automatic Feedback Learning from the recorded video to minimize the errors.
Unsupervised Relation Extraction from Web
September 1, 2013 – November 1, 2013
Project aimed for extracting relation tuples of SPO (Subject, Predicate and Object) form, from an unstructured corpus. Dependency graph obtained from the Stanford parser is extended to extract the tuples. A classifier is trained to learn to distinguish trustworthy tuples from the non-trustworthy and to reduce the computational complexity involved in using a parser on each and every sentence. During the query process, given a partially filled tuple, our system will search for possible entries for the missing fields and rank the resulting tuples based on probabilistic measure. This framework includes three key modules: a. Self-Supervised Learner b. Single-Pass Extractor c. Redundancy-Based Assessor
Google chrome extension Tab Switchy
November 1, 2012 – December 1, 2012
This extension has following features - a) It shows all the tabs closed in the current session and reopens them with just one click b) It has the feature of bookmarking several tabs at a time and also clubbing them under a common name(super-tab) c)The pictorial view and complete title of the page is displayed which helps to identify the tab easily in case of many tabs
Cultural Fit Analysis
The candidate's project portfolio is heavily skewed towards Machine Learning, Deep Learning, and Computer Vision, which aligns well with a data-intensive role. However, the target role is 'Data Analyst', which typically requires strong statistical analysis, data visualization, and business intelligence skills, which are not explicitly highlighted in the projects or experience. While the candidate has a strong technical background, the direct alignment with typical Data Analyst responsibilities (e.g., SQL, Tableau, A/B testing, business metrics) is not clearly demonstrated. The breadth of skills is high within ML/AI, but less so for traditional data analysis.
Soft Skills & Operational Fit
The candidate's project descriptions indicate a strong problem-solving aptitude and an ability to tackle complex technical challenges. The roles at Google and Microsoft suggest experience in structured, large-team environments. However, without specific assessment data on communication, logical reasoning, or teamwork, a definitive assessment of soft skills and operational fit is limited.