AI behavior: Ship fast

Hi, I'mShadab Jamadar.

AI Engineer & AI Product Builder

Building AI products so humans can spend more time creating new problems.

Explore My Work

Let's Connect

Focus & Stream

Currently Building & Activity

In Progress

AI Mock Interview Platform

FastAPI and Redis multi-container architecture. Designed and engineered role-based authentication and secure APIs supporting 100+ concurrent candidate sessions.

EXPECTED RELEASE

Q3 2026

DETAILS

Explore Details

Development Progress85%

Recent Activity

Thoughts

"Code is a liability; features are assets. Write less code."

— Shadab Jamadar

Tomorrow's perspective is currently loading.

Experience Timeline

Professional Journey

Milestones and roles in software development, data science, and AI research that have shaped my technical approach.

View Full Journey

Built a unified Python algorithmic trading SDK integrating 3 broker APIs (Zerodha, Angel One, FivePaisa) using a modular adapter architecture.
Designed a broker-agnostic Instrument Master schema and ingestion pipeline in SQLite, normalizing 100% of broker-specific instrument fields into a consistent data model.
Implemented tests covering core trading workflows, added structured logging, and packaged the SDK with Poetry, enabling scalable onboarding of new brokers.

Engineered end-to-end backend for AI Interviewer platform, designing APIs, integrating databases, and implementing security protocols, reducing backend errors by 25%.
Developed role-based authentication (admin, HR, student) with secure OAuth2.0 management, supporting 100+ concurrent users.
Built a multi-container Docker setup with FastAPI and Redis, cutting startup time by 30% and enabling scalable infrastructure.

Led competitive analysis of 4 key industry players (tech stacks, strategy, financials), uncovering 7 innovation gaps, improving Planto AI’s strategic planning effectiveness by 4%.
Identified and resolved performance bottlenecks in core AI solutions, boosting processing speed by 15% and improving market positioning by 12%.
Built research briefs summarising technical and market insights, enabling 15+ data-informed decisions across product and engineering teams.

Featured Work

Flagship Products

A selection of AI products and tools built end-to-end, solving real classroom, developer, and data-engineering challenges.

Education AI

Curio - AI Co-Teacher.

An Agentic AI system that autonomously handles 5+ core teaching tasks, reducing manual workload for teachers by up to 40%. Generates educational assets like quizzes, images, and summaries, and integrates real-time voice interaction using Gemini for live classes.

LangGraphFastAPIPicovoiceGemini APIReact

View Project

AI Engineering & Observability

LLM Evaluation Pipeline.

A deterministic evaluation pipeline measuring relevance, completeness, hallucination, and latency on 150+ queries. Features modular loader-metrics-aggregator workflows, embedding caching, and local inference for 25% faster batch executions.

PythonSentence TransformersRedisDockerFastAPI

View Project

Fintech & Developer Tools

Filoos - Algorithmic Trading SDK.

A unified Python algorithmic trading SDK integrating Zerodha, Angel One, and FivePaisa broker APIs. Designed a broker-agnostic instrument master schema and sqlite ingestion pipeline that normalizes 100% of broker-specific fields.

PythonSQLitePoetryREST APIsStructured Logging

View Project

main.py - filoos

from filoos.brokers import Zerodha

from filoos.instruments import Equity

from filoos.constants import TransactionType

# Initialize broker safely without leaking API strings

broker = Zerodha(api_key, access_token)

# Place order using strict Domain Objects

order_res = broker.place_order(

Equity("RELIANCE"),

TransactionType.BUY,

quantity=10

)

print(f"Order placed: {order_res.order_id}")

TERMINAL

$ python main.py
[INFO] SQLite Resolver: Loaded 112,450 instruments
[INFO] Resolving token for Equity(symbol='RELIANCE') -> 738561
Order placed: 24012500194833

Skills Knowledge Graph