Harsh Poddar

Data Scientist

Uttarakhand, India4 yrs 10 mos experience

AI EnabledAI ML Practitioner

Key Highlights

Reduced forecasting error by 20-30% for clients.
Cut model training time by 75% using AWS SageMaker.
Accelerated OCR workflows by 20x at Quantiphi.

Stackforce AI infers this person is a Data Scientist specializing in Healthcare and AI-driven solutions.

Contact

Skills

Core Skills

Data ScienceMachine Learning

Other Skills

AI pipelinesAWS SageMakerAWS TextractAirflowAmazon Web Services (AWS)Attention to DetailAutomationCoding ExperienceCommunicationComputer VisionContinuous Integration and Continuous Delivery (CI/CD)Convolutional Neural Networks (CNN)Data AnalysisData StructuresDeep Learning

About

I’m a Data Scientist at Sequoia Consulting with 3.5+ years’ experience in time-series forecasting and generative AI. I help HRTech, healthcare, and finance clients reduce forecasting error by 20–30% and cut model training time by 75%. My work includes building a Generative AI chatbot for healthcare insurance, automating document ingestion pipelines with Airflow, and migrating headcount forecasting to AWS SageMaker. Previously at Quantiphi Analytics, I accelerated OCR workflows 20× and developed diffusion models for synthetic image generation. I’m proficient in Python, PyTorch, TensorFlow, AWS SageMaker, GCP Vertex AI, Docker, Kubernetes, and Airflow, and I hold AWS and GCP associate certifications. With a B.Tech in Biotechnology from IIT Roorkee, I’m driven by continuous learning and cross-functional collaboration to deliver measurable ML solutions. Let’s connect if you’re looking to optimize forecasting or scale generative AI in your organization.

Experience

4 yrs 10 mos

Total Experience

2 yrs 6 mos

Average Tenure

2 yrs 4 mos

Current Experience

Sequoia

Data Scientist

Feb 2024 – Present · 2 yrs 4 mos · Bengaluru, Karnataka, India · Hybrid

PythonAWS SageMakerGenerative AITime-series forecastingModel trainingData Science+1

Quantiphi

3 roles

Machine Learning Researcher

Promoted

Mar 2023 – Feb 2024 · 11 mos

Employed in the Lifescience and Healthcare division of Quantiphi's R&D unit.
Focused on a Generative AI use case in proteomics, utilizing diverse architectures, including Diffusion and Transformers.
Orchestrated end-to-end pipelines integrating multiple AI models and an array of bioinformatics tools.

Generative AIDiffusion modelsProteomicsAI pipelinesMachine Learning

Machine Learning Engineer

Promoted

Dec 2021 – Mar 2023 · 1 yr 3 mos

Worked on end-to-end SOTA Diffusion model development from Research, Data Curation, Model Development, and ML Ops.
Experienced with cloud services for model development and Data Processing.
Completed AWS-certified Solution Architect Associate and GCP Associate Cloud Engineer Certification.
Worked on the use case of OCR technique using AWS Textract and Detectron model and achieved 85% accuracy.

Diffusion modelsOCR techniquesAWS TextractDetectronMachine Learning

Machine Learning Engineer

Aug 2021 – Dec 2021 · 4 mos

Build a vanilla Neural Network model from scratch using NumPy library.
Learned and applied multiple Neural Network Architecture such as VGG16, ResNet50, InceptionV3 on image datasets.
Learning RNN, Transformer, Attention, BERT, GPT model for Neural Language Processing tasks.

Neural NetworksNumPyImage datasetsMachine Learning