Saraswathi Kilaru

Over a 4-month internship, I worked on building and optimizing an AI-powered learning and role-play system using Large Language Models (LLMs). The goal was to design intelligent conversational agents that simulate real-world scenarios and provide structured feedback to users. I was responsible for designing and implementing prompt-based AI workflows using LLM APIs (such as OpenAI models), focusing on improving response quality, consistency, and contextual understanding across multi-turn conversations.

Promoted to Team Lead role in the same AI focused project, leading development efforts to build AI driven backend systems using Python and FastAPI. I contributed to designing scalable architectures on Microsoft Azure and integration of LLM APIs for building intelligent features. I was responsible for building RAG pipelines with document processing, semantic chunking and embedding based retrieval systems. Furthermore, I developed agent-based workflows for multi-step task execution and developed evaluation fr

Program Analyst specialising in Python development and early-stage AI exposure. Backend automation, data processing, and system integration tasks using Python and FastAPI. I contributed to the development of REST APIs and the fixing of internal engineering workflows. Besides development work, I explored the basics of Machine Learning like data analysis, feature engineering and basic model building using Numpy, Pandas and Scikit-learn. I also learnt about Git, CI/CD pipelines and the basics of Docker. In thi

Upon graduating in engineering, I launched my professional career as a work-study in Python development, applying my software engineering fundamentals to real-world projects. My main responsibilities included writing performant python scripts for backend automation, data processing and system integration tasks. I worked on building and maintaining small scale backend services, improving code performance and supporting application workflows. I started to pivot into AI slowly looking into machine learning co

In my last year engineering thesis project I have worked on the image security enhancement using the steganalytic methods. The project was designed to detect hidden data in digital images and enhance data security against steganography attacks. I created a Python system for analysing image features using computer vision and statistical techniques to detect anomalies that could represent hidden data. The project included preprocessing of images, feature extraction and application of basic machine learning mo