Scroll

Hi, I'm

Applied AI Engineer

I build practical AI systems that work beyond the demo.

RAG, agentic workflows, voice and real-time applications — built with a production mindset.

SM

A concise spoken introduction with natural gestures.

01 / Selected work

Work that reached users.

Systems I have designed, built and shipped — chosen for the problems they solve, not the frameworks they use.

Flagship

Remote cervical screening, live

VisiSpec — Real-time clinical AI

Real-time WebSocket video linking remote screening sites to clinicians, inside an AI-enabled cervical-screening device.

  • Python
  • FastAPI
  • WebSockets
  • Real-time video
Outcome — Published in the International Journal of Gynecological Cancer.Read the publication

Multi-agent voice interviews

Agentic hiring workflow

LangGraph agents parse CVs, match them to job descriptions and run live voice interviews via the OpenAI Realtime API.

  • LangGraph
  • OpenAI Realtime API
  • Python
  • FastAPI
Outcome — End-to-end screening without a human in the first loop.

Retrieval that answers the phone

RAG appointment concierge

FastAPI RAG service that verifies hospital appointments conversationally, grounded in live scheduling data.

  • RAG
  • FastAPI
  • PostgreSQL
  • Redis
Outcome — Cut manual call volume by 40%.

Conversational AI on web + WhatsApp

Maternity care assistant

LangChain assistant answering maternity patients on web and WhatsApp with consistent, safe guidance.

  • LangChain
  • WhatsApp API
  • Python
Outcome — Lifted engagement by 35%.

VS Code extension, published

Grove — parallel AI development

VS Code extension launching Claude Code agent teams in isolated Git worktrees, with conflict prediction and guided merges.

  • TypeScript
  • VS Code API
  • Git
  • Python
Outcome — Published on the VS Code Marketplace.View on GitHub

02 / About

I work where machine learning meets the messy real world hospitals, hiring pipelines, screening devices. My systems verify appointments over the phone, interview candidates by voice, and connect remote clinics to specialists in real time. Not demos. Deployed software that people depend on.

0+

years shipping applied AI

0%

manual call volume removed

0%

engagement lift, conversational AI

0

published research papers

03 / Experience

Where I have shipped.

May 2023 — Present

AI Engineer

Odyssey Therapeia — Bengaluru, India

Core AI engineer for a healthcare technology company — owning conversational, retrieval and real-time systems end to end.

  • Engineered a FastAPI-based RAG concierge for appointment verification, reducing manual call volume by 40%.
  • Built a LangGraph multi-agent hiring workflow with the OpenAI Realtime API — CV parsing, JD matching and live voice interviews.
  • Developed a LangChain conversational AI across web and WhatsApp for a maternity hospital, lifting user engagement by 35%.
  • Implemented the real-time WebSocket video workflow connecting remote screening sites to clinicians for VisiSpec, an AI-enabled cervical-screening device.

Feb 2023 — May 2023

Software Developer Intern

Odyssey Therapeia — Bengaluru, India

Neurotech tooling for research teams.

  • Built a React + Chart.js dashboard on the Neurosity Crown SDK to visualise live EEG data for neurological research.
  • Accelerated researcher analysis workflows by 60% with interactive brain-activity visualisations.

04 / Skills

A stack I can defend.

A deliberately lean stack — every tool here has shipped to production at least once.

01

LLM & agent systemsThe reasoning layer

  • RAG
  • LangGraph
  • LangChain
  • OpenAI Realtime API
  • Prompt engineering
  • MCP
02

Languages & backendThe delivery layer

  • Python
  • TypeScript
  • SQL
  • FastAPI
  • REST APIs
  • WebSockets
03

Data & retrievalThe memory layer

  • PostgreSQL
  • Redis
  • Pinecone
  • Chroma
  • Embeddings
  • Vector search
04

Cloud & deliveryThe shipping layer

  • AWS
  • GCP
  • Docker
  • Kubernetes
  • CI/CD
  • Git

05 / Research & writing

Peer-reviewed, not just deployed.

International Journal of Gynecological Cancer, Feb 2026

VisiSpec — an AI-enabled vaginal speculum integrating coaxial illumination and intra-blade imaging for cervical screening

Shebin Mohan Kodumthara (co-author). International Journal of Gynecological Cancer 36(2 Suppl 1):103307, Feb 2026 — conference abstract.

DOI: 10.1016/j.ijgc.2025.103307

International Journal of Novel Research and Development, June 2023

Automated Arm Robot for Intelligent Surveillance by Integrating Image Processing

Shebin Mohan K (co-author). International Journal of Novel Research and Development 8(6):b274–b279, June 2023 — final-year project.

Paper: IJNRD2306129

06 / Contact

Let's buildsomething thatships.

I'm open to Applied AI Engineer roles in the UK and remote-friendly teams anywhere. If you're building products where AI has to actually work, I'd like to hear about it.