Vikas Vikku

Detail-oriented Software Engineer with over 3+ years of hands-on experience in Data Science, Machine Learning, and Full-Stack Development. Proficient in writing high-quality, maintainable Python code using OOP, SOLID principles, and design patterns. Skilled in developing RESTful APIs with FastAPI, integrating ML/DL models, and implementing CI/CD pipelines with Docker and Git.

Experienced in MLOps practices using MLflow and DVC, with expertise in cloud platforms like AWS. Adept at collaborating with cross-functional teams to deliver scalable, efficient solutions with a focus on performance monitoring and optimization.

AI Developer - CIGI Technologies - Bangalore, Karnataka

(2025-12 - 2026-09)

• Designed and implemented xolaris-ai, a suite of Python-based AWS Lambda agents for automating accounting, spend governance, forecasting, and reconciliation processes.
• Automated invoice capture and expense extraction by integrating QuickBooks APIs and AWS Textract, improving financial data accuracy and operational efficiency.
• Developed advanced forecasting and pattern learning modules leveraging AWS Bedrock, AWS S3, and AWS DynamoDB to enhance financial prediction capabilities.
• Built and maintained RESTful APIs and automated workflows to streamline enterprise financial operations and reporting.
• Ensured code quality and deployment reliability through comprehensive test suites, documentation, and CI/CD best practices.
• Developed AI/ML-powered financial automation solutions using Python, AWS Bedrock, forecasting models and GenAI techniques.
• Built AI agents for accounting, spend governance, forecasting and reconciliation processes.

Software Developer - Cyphertek Solutions Pvt Ltd - Bangalore, Karnataka

(2025-06 - 2025-12)

• Developed a real-time deep learning model using YOLOv3 and PyTorch for object detection, achieving 96% accuracy across 61 error classes in manufacturing error detection.
• Designed and implemented RESTful APIs using FastAPI for model serving, enabling seamless integration with production systems.
• Developed and deployed a deep learning-based computer vision solution using YOLOv3 and PyTorch for manufacturing error detection.
• Implemented MLflow for experiment tracking and model versioning following MLOps practices.
• Utilized Docker for containerized deployment on the RZ-AI/V2M board, ensuring scalability and reliability.
• Implemented MLflow for experiment tracking and model versioning, adhering to MLOps best practices.
• Incorporated structured logging and performance monitoring to ensure system observability and robustness.

Software Engineering Intern - WiseWork - Bangalore, Karnataka

(2024-07 - 2025-03)

• Built a topic extraction system using OpenAI API and FastAPI, improving data processing accuracy by 25%.
• Integrated Qdrant vector database for efficient storage and retrieval, reducing query times by 40%.
• Optimized LLM prompts using ell library, reducing API costs by 15% while enhancing response quality.
• Developed a ContentService to process multiple document formats (PDF, DOCX, HTML), supporting over 10 formats.
• Leveraged SpaCy (en_core_web_trf) for advanced NLP tasks, improving topic extraction accuracy by 20%.
• Implemented CI/CD pipelines using Jenkins and Git, ensuring repeatable deployments.
• Contributed to Wise Maestro, a multimodal RAG application, handling both frontend and backend development.

Quantity Surveyor - JMC Projects Ltd. - Bangalore, Karnataka

(2020-02 - 2023-07)

• Prepared quantity take-offs, BOQs and cost estimates.
• Managed subcontractor billing, measurements and payment certifications.
• Monitored project quantities, costs and variations.
• Prepared and reviewed client/subcontractor invoices.
• Coordinated with project, engineering and procurement teams.
• Supported cost control and commercial documentation throughout project execution.

Bachelor of Technology - Civil Engineering - Reva University (2015 - 2019)

PGP - Data Science and Machine Learning - MITxMicroMasters and Intellipaat (2023 - 2023)