11/15/2026 updated

**** ******** ****
100 % available

Aspiring AI Engineer | Machine Learning Specialist

Pune, India
Worldwide
B.E - Artificial Intelligence and Data Science
Pune, India
Worldwide
B.E - Artificial Intelligence and Data Science

Profile attachments

Soham_Jadhav_Resume.pdf

AI AgentsLLMArtificial IntelligenceComputer VisionPython (Programming Language)Machine LearningTensorFlowFeature EngineeringData SciencePyTorchMulti-Agent SystemsDeep LearningGenerative AI
TECHNICAL SKILLS
Languages & Core Frameworks: Python, C++, SQL, PyTorch, TensorFlow, OpenCV, FastAPI, Docker, API, Git/CI-CD
Generative & Agentic AI: LLM, LangChain, RAG, Multi-Agent Workflows, LoRA/QLoRA, Transformers, Fine-Tuning, Automations
ML Systems & Computer Vision: U-Net, CNNs, BiLSTM, Computer Vision, Segmentation, Modular Pipelines and Architecture


PROFESSIONAL EXPERIENCE

Google Summer of Code 2026 |Machine Learning for Science (ML4Sci)
Research & Development | PREDICT1
Developing reproducible CAC segmentation benchmarking pipelines on the Stanford COCA dataset to systematically evaluate deep learning architectures for clinical-scale cardiovascular imaging.
Benchmarked U-Net against TotalSegmentator, achieving a 0.9416 Dice score and ~63x faster inference efficiency on targeted CAC diagnostic workflows.

Google Developer Groups (GDG) On Campus, Nashik
Co-Lead | Artificial Intelligence & Machine Learning
Led the AI & ML vertical, architecting and delivering 10+ hands-on technical workshops on Deep Learning, RAG, and LangChain orchestration for an audience of 80+ engineering students.
Provided direct mentorship on PyTorch and HuggingFace workflows; awarded Certificate of Appreciation for technical leadership and community contribution.


PROJECTS

Convo-Ease: Intelligent Multi-Modal Content Moderation
Gemma 3 27B, BLIP-2, Whisper Large-v3, LangChain, FastAPI
Architected a pre-delivery Gatekeeper framework validating text, image, and audio in under 3 seconds using dynamic Policy-as-Prompt logic; published in Cureus, Springer Nature 2026.

PrediCT Studio: Auditable Coronary Calcium Scoring Workstation
Python, FastAPI, PyTorch, MONAI, nnU-Net, TotalSegmentator, Three.js
Built a multi-user clinical workstation (FastAPI, Three.js) that scores coronary calcium from cardiac CT and lets clinicians audit every lesion; its soft-coverage U-Net cuts median Agatston error 2.2x (42.97 to 19.27).

Copilot for Data Science and Analysis
Python, NLP, AutoML, XAI, RAG, LangChain
Automated end-to-end EDA, feature engineering, and model selection via natural-language execution, reducing manual pipeline overhead by 90% while producing explainable outputs.

RenAIssance OCR: Historical Document Recognition
PyTorch, ResNet-18, BiLSTM, CTC Loss
Engineered a domain-specific CRNN pipeline with LLM post-processing to transcribe Renaissance-era manuscripts, restricting outputs to historically valid vocabulary.


EDUCATION
B.E. in Artificial Intelligence and Data Science
MET's Institute of Engineering, Nashik (Affiliated to SPPU)

Experience

Professional experience0 - 5 years
Freelancer since2024

Languages

GermanBasic knowledgeEnglishFluentFrenchBasic knowledgeHindiFluent

Project history

Research and Development

Machine Learning For Science
Developed reproducible CAC segmentation benchmarking pipelines on the Stanford COCA dataset to systematically evaluate deep learning architectures for clinical-scale cardiovascular imaging.
Benchmarked U-Net against TotalSegmentator, achieving a 0.9416 Dice score and ~63x faster inference efficiency on targeted CAC diagnostic workflows.

Co-Lead | Artificial Intelligence and Machine Learning

Google Developers Groups (GDG) On Campus, Nashik
Engaged students in learning about AI, Machine Learning, and Generative AI. Taught and trained peers on AI and technology.

Contact form

Log in to get in touch

You need to be logged in to use the contact form.

Sign upLog in