open to AI engineering roles

I build AI thatdoes the work.

Full-stack AI developer in Delhi. I ship agentic systems, RAG pipelines and the FastAPI / Next.js machinery around them — the kind that survive contact with production, not just a demo notebook.

See the work Résumé Get in touch

rama@portfolio — ~/whoami

$whoami --verbose

name: Ramachokkalingam S A
role: Full-Stack AI Developer
focus: agentic AI · RAG · cloud AI
stack: LangGraph · FastAPI · Next.js
edge: ROS · SLAM · Jetson Nano
base: Delhi, India
langs: ta · en · hi · te · kn

try

70–95%manual business effort removed through AI automation

15+national innovation competitions reached as finalist

400+students taught across AI & IoT sessions

10AI and robotics systems designed end to end

about

Translator between
humans and machines.

I'm Ramachokkalingam S A — Rama is fine. I build systems where the model actually does something: LangGraph agents that decompose a task and hand off to each other, RAG pipelines that retrieve the right chunk instead of a plausible one, and the FastAPI and Next.js scaffolding that turns both into something a team can depend on.

Most recently I was a Data Analyst Associate at TURMS Intelligent Apparel, modelling warehouse data in BigQuery and dbt and shipping Power BI dashboards operations opened every morning. Before the LLM stack I came up through robotics — SLAM rovers, YOLO on Jetson Nano, a drone that watches water for drowning swimmers. That's why I care about latency, sensors and the failure modes that only show up outside the notebook.

Off-hours I fly drones for photography, judge hackathons, run the technical club, and teach anyone who'll sit still long enough.

currently

Integrated B.Tech + M.Tech in AI & ML, Amity University Gurugram
President, TECHNIKI Technical Club · AI & ML Lead, Microsoft Student Ambassadors
Building agentic tooling — multi-agent loops, MCP servers, evaluation harnesses
Building AmiBot — a humanoid on ROS2, from the SolidWorks chassis to off-board inference
Open to full-time AI / ML engineering roles

certifications

✓ GitHub Foundations
✓ AWS Academy — Generative AI Foundations

experience

Where the hours went.

Three years of shipping — analytics platforms, university web infrastructure, edge AI and the classrooms in between.

Jun 2025 – Dec 2025Bengaluru

Data Analyst Associate

TURMS Intelligent Apparel

Built Power BI dashboards tracking warehouse KPIs — inventory, order fulfilment and supplier performance.
Modelled WMS data in BigQuery with dbt, creating the structured reporting layer the business now runs on.
Processed high-volume warehouse data with PySpark and Kafka; automated ingestion through REST APIs into PostgreSQL.
Ran EDA, statistical analysis and ML for demand forecasting and anomaly detection with Pandas and scikit-learn.

Aug 2024 – May 2025Gurugram

Webmaster

Amity University

Ran high-traffic university web platforms on Node.js and React.
Cut page load times by 40% and moved releases onto GitHub Actions CI/CD.

Jun 2024 – Jul 2024Bengaluru

AI & IoT Intern

YoLabs

Built IoT systems on Raspberry Pi wired into cloud APIs.
Worked with CNN vision models and ROS robotics workflows; delivered AI workshops to student cohorts.

2022 – presentGurugram

Integrated B.Tech + M.Tech, AI & ML

Amity University

President, TECHNIKI Technical Club · Tech Lead, Entrepreneurship Cell · Core team, Amifest.
AI & ML Lead, Microsoft Student Ambassadors.

featured work

Three systems worth
a closer look.

Agent orchestration, an agentic terminal, and retrieval that holds up under real questions.

agentic

Multi-Agent AI Workflow System ↗

LangGraph · FastAPI

Multi-agent orchestration that decomposes a task, retrieves documents, searches the web and writes the report — exposed over FastAPI with real-time streaming.

LangGraph
FastAPI
Next.js
pgvector
Redis
Kubernetes

Why: One model answering a research question in a single pass gets shallow answers. Splitting the job across agents that each do one thing well gets a better one.
How it works: A planner breaks the question down, a retriever and a web searcher run in parallel, a synthesiser writes the draft, and a verifier checks it before the report goes out — the whole loop is a LangGraph state machine.
Notable: It ships with the unglamorous parts: a worker queue, pgvector for embeddings, Kubernetes manifests, and Prometheus/OpenTelemetry so you can see where a run spent its time.

agentic

CMDR ↗

Agentic AI terminal · browser-native · built on ttyd

A terminal that runs in the browser — persistent sessions, an auto-update channel, a FastAPI cloud backend and an AI command bar that writes the command for you.

C
xterm.js
FastAPI
React
Docker
PostgreSQL

Why: A terminal you can open from any machine, that remembers where you left off, and that can turn a sentence into the right command.
How it works: The core is built on ttyd, extended with persistent sessions and an auto-update channel. A FastAPI service handles accounts and history; the front end is React and TypeScript over xterm.js with WebGL rendering.
Notable: The AI command bar is Gemini-backed — you describe what you want, it drafts the command, you decide whether to run it.

rag

RAG Chatbot ↗

FastAPI + Next.js full-stack

Production-ready retrieval chatbot: FastAPI backend, Next.js front end, ChromaDB/FAISS vector search, shipped to AWS behind Docker and Terraform.

FastAPI
Next.js
LangChain
ChromaDB
FAISS
AWS

Why: Most retrieval demos answer confidently and cite nothing. This one hands back the passage it used, so you can check it.
How it works: Documents are chunked and embedded into Chroma or FAISS; a FastAPI service does the retrieval and calls the model; a Next.js front end streams the answer with its citations attached.
Notable: Deployed rather than demoed — Terraform stands up the AWS ECS infrastructure, and the whole thing runs in Docker.

All 10 projects

robot bench

Four joints,
live in your browser.

This is the arm geometry I work with in ROS2 — base yaw, shoulder, elbow, wrist and a two-finger gripper. Move your pointer and it tracks you; pick a named pose and it runs it. The numbers below are the angles actually going to the joints.

joint statemode · pointer

base yaw+0.0°

shoulder pitch+0.0°

elbow pitch+0.0°

wrist pitch+0.0°

gripper grip+0.0°

The terminal at the top drives it too — try robot pick.

4-axis arm · rendered live base shoulder elbow wrist gripper

stack

The toolbox.

Eight domains, one throughline: getting a model from a research idea into something that answers a request in under a second.

01 Agentic AI & GenAI

LangGraph
LangChain
LlamaIndex
Model Context Protocol
RAG pipelines
Multi-agent frameworks
Prompt optimisation
LoRA / QLoRA

02 Backend & APIs

FastAPI
Flask
Node.js
Express.js
REST architecture
PostgreSQL
MySQL
MongoDB
Docker

03 Frontend & Delivery

Next.js
React
TypeScript
JavaScript
Dart / Flutter
Vercel
Render
GitHub Actions
CI/CD

04 Data & Analytics

BigQuery
dbt
PySpark
Kafka
Snowflake
Pandas
NumPy
Power BI
SQL
scikit-learn

05 Vision, NLP & Deep Learning

PyTorch
TensorFlow
Transformers
YOLO
SAM
DINO
OpenCV
OCR
Multilingual NLP
Speech synthesis

06 Cloud & MLOps

AWS
Azure AI
Google Cloud
SageMaker
Vertex AI
Kubernetes
MLflow
Terraform

07 Vector & Retrieval

ChromaDB
Qdrant
FAISS
Pinecone
Semantic retrieval
ELT workflows

08 Robotics & Edge AI

ROS
SLAM
LiDAR
UAV control
Jetson Nano
Raspberry Pi
Sensor fusion

selected wins

Beyond the repo.

achievements

Finalist in over 15 national innovation competitions.
President of TECHNIKI; AI & ML Lead for Microsoft Student Ambassadors.
Orchestrated national-level technical hackathons and expert workshops.
Instructed AI & IoT sessions for a cohort of 400+ students.
Invited technical judge at the Control+Think Innovation Hackathon.

languages

Tamil · English · Hindi · Telugu · Kannada

elsewhere

rama@portfolio — ~/contact

$./say-hello --to rama

Got something worth
building? Let's talk.

Agentic systems, retrieval that behaves, data platforms, or a robot that needs a brain — if it's an interesting problem I'd like to hear about it. Fastest route is email.