Tanmay Vaghale

AI Researcher

Pune, Maharashtra, India2 yrs 1 mo experience

Most Likely To Switch

Key Highlights

Expert in Machine Learning and Data Science.
Developed advanced pathfinding visualization tools.
Achieved high accuracy in Named Entity Recognition.

Stackforce AI infers this person is a Data Scientist with strong expertise in Machine Learning and Natural Language Processing.

Contact

Skills

Core Skills

Machine LearningData ScienceSoftware EngineeringTestingData StructuresAlgorithm DesignNatural Language Processing

Other Skills

A-star AlgorithmAutomation ToolsBFSBi-LSTMC++CRFData AnalysisData VisualizationDeep LearningDijkstra's AlgorithmGraph TheoryHealeniumMatplotlibMicrosoft OfficeNamed Entity Recognition

Experience

Turing

Delivery Data Scientist

May 2024 – Present · 1 yr 10 mos · Pune, Maharashtra, India · Remote

Worked on Supervised Fine Tuning (SFT) and Reinforced Learning with Human Feedback for various LLMs
Validated the response of Meta Llama 4 by providing PostgreSQL-related prompts, ensuring alignment with expected performance
Contributed to the creation of more than 4000 complex charts for the SFT of Google Gemini, enhancing data visualization for better model performance

Supervised Fine TuningReinforced LearningPostgreSQLData VisualizationMachine LearningData Science

Honeywell

Software Engineering Intern

May 2023 – Jul 2023 · 2 mos · Bengaluru, Karnataka, India · On-site

Strategised with the testing and automation team to implement self-healing in the test framework
Researched 10+ available automation tools and selected Healenium as the preferred tool for self-healing
Conducted an assessment of the self-healing tool's features and compatibility on a prototype website
Integrated Healenium with 80% confidence into building management test automation framework
Drafted a step-by-step Healenium guide for seamless adoption by the testing and automation team

Self-healingAutomation ToolsHealeniumSoftware EngineeringTesting

Indian institute of technology, kharagpur

2 roles

Path Finding Visualization Tool

Feb 2023 – Mar 2023 · 1 mo

Implemented A-star, BFS, and Dijkstra's algorithms to find the shortest path on a 2D grid with obstacles
Developed a dynamic 2D grid using pygame, enabling users to choose start, end, and obstacles
Implemented an indexed priority queue in Dijkstra's Algorithm, updating distances in O(logN) time
Project link: https://github.com/Tanmay-Vaghale/Path-Finding-Visualization-tool

A-star AlgorithmBFSDijkstra's AlgorithmpygameData StructuresAlgorithm Design

Bachelor Thesis Project | Improving Named Entity Recognition with Word Segmentation

Aug 2022 – Nov 2022 · 3 mos · On-site

Supervisor: Prof. Pawan Goyal | Department of Computer Science and Engineering | IIT Kharagpur
Applied joint Bi-LSTM and CRF training on Bakeoff-4 dataset for Named Entity Recognition
Used the negative log-likelihood of label sequences across all training samples as the loss function
Achieved 99.44% validation accuracy in 30 epochs with 5% random data sampling for training and testing

Bi-LSTMCRFNamed Entity RecognitionNatural Language ProcessingMachine Learning