SecureDocAI
An offline-first document intelligence system for retrieval, summarization, and evidence-aware question answering.
Architecture Signal
Document
Parsing
OCR
Chunking
Embeddings
FAISS
Retrieval
Offline LLM
Answer

Builder • Problem Solver • Storyteller
Focus areas
4
AI systems, web products, automation, creative tech
Featured projects
3
Core work highlighted with deep case-study intent
Current mode
BUILD
Designing, shipping, and iterating in public
TECH STACK
Technologies I use to turn ideas into intelligent systems, products, and experiences.
Core language I use for applied AI systems, automation, and data-heavy engineering work.
0124
SYSTEM IN CONTEXT
How Python operates in my production workflow
From AI pipelines to polished interfaces, I work at the intersection of systems thinking and creative execution.
I design intelligent systems around retrieval, reasoning, evaluation, and reliable outputs — from document intelligence and RAG pipelines to LLM-powered workflows and agentic systems.
System Flow
DOCUMENTS
RETRIEVAL
CONTEXT
LLM
ANSWER
A focused set of work that reflects system-level thinking and practical implementation discipline.
An offline-first document intelligence system for retrieval, summarization, and evidence-aware question answering.
Architecture Signal
Document
Parsing
OCR
Chunking
Embeddings
FAISS
Retrieval
Offline LLM
Answer
A full-stack attendance management platform with OTP-based verification, session-aware validation, analytics dashboards, and CSV export workflows.
Architecture Signal
Student
Class Session
OTP Request
OTP Verify
Attendance Mark
Analytics
CSV Export
An AI-powered conversational commerce platform that translates natural-language shopping intent into structured product retrieval and recommendations.
Architecture Signal
Query
Pydantic AI Agent
Filter Extraction
Normalization
Mongo Query
Product Search
Recommendations
Results
Engineering depth, creative instinct, and an obsession with building useful things.
I care about systems that are technically sound and emotionally legible. My work sits at the intersection of AI engineering, practical product building, and narrative thinking.
I approach projects by identifying the real user friction first, then designing architecture and interfaces that reduce complexity instead of hiding it.
Recent roles and project contexts that shaped how I ship robust systems.
May 2026 — August 2026
Siemens • Pune, India
Built practical enterprise experience across dashboard engineering, data visualization, automation, and AI-assisted workflows, contributing to more reusable, interactive, and efficient analytics solutions.
2025
Delhi Metro Rail Corporation (DMRC) • India
Provided valuable early exposure to real-world engineering and transportation operations, helping connect my academic foundation with practical industry environments.
An evolving body of work around private-by-default document intelligence and practical evaluation methods.
SecureDocAI Pipeline Preview
TODO: Add validated quantitative metrics only after evaluation runs are finalized.
┌─ AJITESH SYSTEM MONITOR ───────────────┐
CURRENT PROCESS
────────────────────────
> building intelligent systems
ACTIVE MODULES
────────────────────────
LOCATION
INDIA / UTC+5:30
STATUS
████████████████ READY
VISUAL BRAND INDICATORS
Open for thoughtful collaboration
If your team is working on AI-heavy products, automation strategy, or narrative-driven digital experiences, I'd love to hear what you're solving.