Headline

AI Analyst @ Augnito || AI undergrad @NITK26

About

I am a passionate Artificial Intelligence student at the National Institute of Technology Karnataka, Surathkal, with a strong foundation in Machine Learning, Deep Learning, and Natural Language Processing, with hands-on experience in AI engineering and R&D internships. I’ve worked on scalable voice agent systems, advanced NLP for document analysis, and developed an advanced Retrieval-Augmented Generation (RAG) system for ESG documents. My projects include bankruptcy prediction using NLP and urban mobility management. Skilled in Python, Java, and machine learning frameworks, I’m also a student-athlete and active participant in campus organizations and events. I thrive on solving complex problems and transforming unstructured data into actionable insights. With a keen interest in AI research and its real-world applications, I am eager to contribute to innovative projects and collaborate with like-minded professionals in the tech industry.

Location

Greater Bengaluru Area

AI stack

opencode

claudecode

Work experience

AI Analyst at Augnito, Jun 2026 - Present · 3 mos, Hybrid

NLP Research Intern at Augnito, Jan 2026 - Jun 2026 · 6 mos, On-site, • Conducted R&D on drug spell correction pipelines for clinical vocabulary normalization in medical transcription workflows. • Built an LLM-as-Judge framework to assist SOAP validation, boosting throughput by 10× with metrics-driven quality checks. • Contributing to multiple confidential R&D initiatives in healthcare LLM adaptation under restricted information-sharing guidelines. • Co-authored a research paper on clinical LLM evaluation for medical hallucination detection: arxiv.org/abs/2604.14829

AI & Data Engineering Intern at OmicsBank, Aug 2025 - Oct 2025 · 3 mos, New York, United States, - Engineered automated clinical longitudinal data extraction pipelines in Python, processing any heterogeneous patient records. - Built batch processing workflows & semi-automated evaluation pipeline, structuring outputs into OMOP Common Data Model. - Achieved 85% extraction accuracy (89% precision, 92% recall) through iterative evaluation & post-processing pipelines.

AI Engineer Intern at Turbostart, Feb 2025 - Jul 2025 · 6 mos, Bengaluru, Karnataka, India, worked as an Junior AI engineering Developer in AI Voice-Agent engineering projects • Developed and deployed a scalable, AI-powered voice agent using Plivo and Ultravox for automated inbound and outbound calls. • Implemented a distributed system using Google Cloud Pub/Sub to manage high-volume asynchronous call tasks. • Built a FastAPI-based webhook service in Python to process call events, update MongoDB, and handle business logic. • Designed a shared state management system using MongoDB for consistent data tracking across application instances. • Performed R&D to build an in-house speech-to-text and text-to-speech service using Fixie-ai STT and Kokkoro TTS, aiming to reduce external API dependency and call starting delay by 1 second.

Software Research Engineer - Intern at Jump1K, Dec 2024 - Feb 2025 · 3 mos, Remote, • Researched various methods for extracting text, tables, and images from ESG documents, including pdfplumber, PyMuPDF, and layout-based detection with Detectron2. • Implemented and evaluated OCR techniques using Mistral AI and Tesseract for handling scanned and low-quality PDFs. • Built a Retrieval-Augmented Generation (RAG) system for ESG document analysis using semantic chunking, query classification, and Google’s GenAI APIs. • Integrated FAISS for efficient vector similarity search and implemented streaming responses with dynamic prompting techniques.

Executive member at IE Robotics, Jan 2023 - Jan 2025 · 2 yrs 1 mo, NITK

Member at IE Robotics, Dec 2022 - Jan 2025 · 2 yrs 2 mos, NITK

Education

Bachelor of Technology - BTech, Artificial Intelligence, National Institute of Technology Karnataka

Inter college, PCMC, Tapas

Public links

github: https://github.com/saichiranthan