Raajas Sode

1. Generative AI Application Development: Designed, developed, and maintained full- stack generative AI applications, covering the entire model lifecycle from training to inference.

2. Large Language Models: Extensive hands-on experience with fine-tuning, evaluating, and deploying client-facing LLMs, including using open-source models like Mistral and Llama and using frameworks like Langchain, LlamaIndex and Huggingface.

3. Python & Bash Proficiency: Developed automation scripts in Python and Bash for deployment, monitoring, and scaling of AI models and infrastructure components.

4. GPU Optimization: Applied deep expertise in GPU-accelerated computing principles to maximize model performance and efficiency during training and inference phases.

5. Retrieval-Augmented Generation (RAG) Systems: Built end-to-end RAG systems using LangChain to generate LLM outputs enriched with curated extended context from a vast corpus of specialized domain data.

6. Building Inference Engines: Designed prompt engineering frameworks optimized for ISO standards and implemented context management systems to maintaining the coherence of policy clauses.

7. Created validation systems based on ISO guidelines and Rubrics for quality assurance of generated context and ensuring it adheres to industry standards.

1. Data Modeling with dbt: Extensive use of dbt for data modeling and transformation within Snowflake.
2. Machine Learning with Python: Developed machine learning models using Python for tasks such as predicting product defects.

3. Data Modelling: Used Snowflake as a Central Data Warehouse to design and implement data infrastructure designs around high quality diverse data sources.

4. CI/CD Pipelines: Implemented CI/CD pipelines using Docker, Kubernetes, and Argo CD for automated deployment and scaling.

5. Data driven Insight: Developed data driven insights using reporting tools like Looker to organise accurate and action oriented dashboards for stakeholders.

6. Machine Learning: Developed machine learning models to predict product defects and writing root cause analysis models that improved efficiency by 30%.

1. Data Warehousing & Pipelines: Created data warehousing infrastructure and pipelines, leveraging Snowflake, to deliver marketing analysis and drive client sales.

2. dbt Model Development: Extensive experience in dbt model development for data transformation and analysis within Snowflake.

3. Tableau Dashboard Management: Managed and maintained Tableau dashboards for data visualization and reporting.

4. Multi-Touch Attribution Modeling: Developed Markov chain models in Python for multi-touch attribution marketing, extracting channel-specific revenue impacts.

5. Python-based Data Integration: Developed data integration pipelines for Python-based machine learning algorithms.

6. ETL Collaboration: Collaborated with stakeholders and marketing experts to understand the data better, which helped ups build better data pipelines and dashboards.