Prakhar Prakarsh
Hi, I'm Prakhar Prakarsh, a Data Engineer with a passion for machine learning, data analysis, and statistical modeling.
About
Hi, I’m Prakhar Prakarsh, a Data Engineer with a passion for machine learning, data analysis, and statistical modeling. With a strong background in Python, SQL, and various data visualization tools, I enjoy turning data into actionable insights that drive business decisions. My experience spans feature engineering, hyperparameter tuning, and building predictive models that balance accuracy with computational efficiency.
Over the past year, I’ve had the opportunity to work on diverse projects ranging from personalized recipe generators at Google to optimizing large-scale Spark ETL pipelines. I’m keen to continue learning, growing, and contributing to innovative data solutions while collaborating with talented teams.
- 3 years' experience
- Joined July 2025
Languages
- EnglishAdvanced
- HindiAdvanced
Industry experience
- Software & Internet
- Professional Services
- Computers & Electronics
- Media & Entertainment
- Education
More experts in Sheffield, United Kingdom
Work experience
Developed a personalised recipe generator by integrating FAISS for vector search and Hugging Face models, achieving 90% semantic relevance across 20,000 recipes. Built a real-time nutritional data lookup system, reducing manual nutritional verification time by 100% through automated enrichment of recipe outputs. Engineered a prompt-driven JSON parser, maintaining 80% valid structured outputs across diverse dietary inputs. Visualised nutritional distributions (calories, protein, fat, sodium) using histograms to support model fine-tuning and user transparency, enhancing model interpretability.
Built and optimised Databricks/Spark ETL pipelines in Python and SQL to process 3 TB of daily data, slashing query times by 40%. Developed and version-controlled dbt models and Terraform modules on AWS, cutting deployment time by 60% and ensuring repeatable infrastructure. Drove CI/CD with BitBucket, GitHub and Azure DevOps shipping 5 analytics features per quarter and reducing production incidents by 30%. Authored and maintained Jira/Confluence docs for data workflows, accelerating team onboarding by 50% and upholding 94.9% platform uptime.
Education
Certifications and awards
Hire Prakhar Prakarsh today
Post a job and invite Prakhar, or contact them directly. No hidden commission.