Building AI that
actually works.
I'm Gowtham D, an AI Engineer passionate about turning complex AI research into production-grade systems that solve real problems. I believe in learning by building — every project I ship teaches me more than any course.
My journey started with Data Science at IIT Madras via GUVI, and evolved quickly into Agentic AI, MLOps, and multi-agent system design. I co-founded Literate Spork, a GitHub community that mentors 50+ students through real-world AI projects.
As AIML Head at GDGoC SKASC, I ran sessions on LLMs, computer vision, and MLOps pipelines — always focused on building, not just learning theory.
My current obsession is Agentic AI — building LangGraph-orchestrated multi-agent systems that can perceive, decide, and act. FHIRFlow is the best example: 5 agents, real FHIR data, live EDI claim submission, and actual AI voice calls to patients.
Gowtham D
AI Engineer · India
JOURNEY
Technical Arsenal
AI & Machine Learning
MLOps & DevOps
Programming Languages
Data Processing & Analytics
Databases & Infrastructure
Cloud Architecture
Roles & Leadership
AI Developer Intern
Orchestrated generative AI systems and multi-agent workflows, driving operational speed and semantic recall optimizations.
- Designed LangGraph multi-agent systems, reducing average LLM system latency by 32% and compute cost by 20%.
- Engineered FastAPI backend serving up to 5,000+ daily production API requests with a 99.9% uptime rate.
- Optimized semantic recall query paths in AWS S3 Vectors , improving search precision by 45%.
AI Developer Intern
Constructed high-speed web scraping, parsing, and classification agents to feed training pipelines for custom LLM models.
- Built automated ingest, parse, and classification scripts using n8n + Python, saving 15+ operational hours per week.
- Ingested and normalized over 2.5 million rows of semi-anonymous web data to train local analytical networks.
- Enhanced local threat classifier performance by 18% through targeted fine-tuning and system prompt engineering.
App & AI Developer Intern
Collaborated with an international engineering team to build, optimize, and scale production backend systems for AI-powered mobile apps.
- Developed and scaled backend APIs for 2 flagship applications (Noulez & SnagNinja) serving thousands of users.
- Optimized SQL indexing and database queries, achieving a 40% reduction in database read response times.
- Integrated push notifications and localized scheduling loops that increased daily active users (DAUs) by 22%.
App & AI Developer Intern
Built a highly optimized machine learning classifier detecting malicious signatures and phishing nodes.
- Engineered an XGBoost/Scikit-learn URL classifier achieving 98.4% accuracy in detecting phishing vectors.
- Curated and normalized a balanced training set of 1.2 million URL signatures to eliminate database bias.
- Containerized security model using FastAPI and Docker, achieving a 95ms median response time on system endpoints.
AIML Head
Directed tech curriculums, built student developer pathways, and mentored coding teams on deep learning systems.
- Led a technical community of 200+ members, conducting 15+ hands-on bootcamps on Deep Learning and MLOps.
- Mentored 10+ student coding teams, resulting in 3 first-place wins at regional and local college hackathons.
- Published 3 comprehensive open-source ML roadmap repositories, gaining 100+ combined student stars on GitHub.
Flutter Head
Educated and trained student engineers in mobile framework architectures, modular coding, and open-source contributions.
- Conducted Flutter bootcamps for 150+ students, resulting in 25+ student-published cross-platform apps.
- Spearheaded collaborative coding seminars, onboarding 120+ student developers to Git and collaborative workflows.
- Architected standard MVVM skeleton codes for chapter projects, ensuring 100% adherence to clean code guidelines.
Founder
Founded and scaling a project-driven GitHub student community providing real-world MLOps mentorship.
- Founded a GitHub collaborative workspace, mentoring 50+ students in building and packaging predictive systems.
- Curated and maintained 20+ open-source project repositories covering NLP, computer vision, and agentic AI.
- Conducted 40+ formal peer code reviews, improving student pull request optimization rates by 25%.
Hackathons
HackNext'26 Series 1.0 @ SNS College of Technology
Built a 17-agent orchestration system for hospitals that reduces patient wait time and panic in medical claims while reducing hospital claim rejection rates by 35%.
ICAITICA 2026 International Conference
Presented research on an AI agent detecting drug trafficking activities on Telegram and Reddit platforms.
Chess Gold @ SKASC Sports Meet 2026
Secured the first place and gold medal in the chess tournament representing Yaayu house at Sri Krishna Arts and Science College.
HackAppsters L&T Edutech Hackathon
Built a deep learning-based medical imaging model to classify 10+ chest diseases from radiograms with high accuracy.
HackVision 2026
Developed a single intelligent agent capable of managing end-to-end event workflows including scheduling, coordination, and automation.
NeoVerse @ CIT 2026
Designed an AI agent to detect illegal activities like drug and human trafficking on semi-anonymous platforms such as Telegram and Reddit.
Datathon @ CIT 2026
Built a multi-agent healthcare system that validates and corrects FHIR claims by comparing them with updated policies.
Kaggle Playground Series
Achieved Top 100 rank twice and Top 50 once in Kaggle Playground Series by building high-performance machine learning models.
Tech Sprint 2026 (GDG SKASC)
Built an autonomous agent that fetches vulnerabilities from NVD and applies backend fixes seamlessly without disrupting users.
SNS Tech Hack 2024
Developed a complete EV charging slot booking application within 24 hours during the hackathon.
Tech Hack '24
AI-based hospital app that connects doctors and patients with intelligent diagnosis and real-time clinical support.