Skip to content
Jaymin Dave
Vol. I — Issue 01Ahmedabad, India — 2026

JayminDave

Backend-first, building AI-powered systems

Available for rolesResume

Ask Jaymin's assistant.

His experience, his projects, how something was built — out loud or in writing.

Press to speak, or type below

Model-generated — verify anything that matters.

01The Long Read

I build the parts of software that don't get screenshots.

I like backend work because it's where a product's logic actually lives — how data moves, how requests get handled, and what happens when something breaks. I usually learn a system by building a small version myself, breaking it, and figuring out why. Most of what I know came from that loop: pick a problem a little harder than the last one, get it working, then go back and make it cleaner.

I care about writing code that's readable and easy to change, and I like working in a structured way — steady, consistent progress rather than last-minute rushes. Lately I've been exploring how AI and LLMs fit into real applications, which keeps reminding me how much there still is to learn. I'm early in my career and comfortable saying so: I'd rather ask questions and understand the reasoning than pretend I have everything figured out.

I build the parts of software that don't get screenshots — the APIs, the queues, the locks that hold under concurrent load.
Enthusiasm
If I had one word for myself, it'd be enthusiastic. I get genuinely absorbed in what I'm building — that focus is what makes me a quick, deep learner.
Adaptability
Comfortable across new tools, teams, and situations. I stay open to learn, unlearn, and relearn as things change.
Problem-solving
I don't stop until it's solved. Once I'm on a problem, it gets my undivided attention until it finally makes sense.
Ownership
Disciplined and reliable. A steady routine keeps me consistent, and people can trust what I take on to actually get done.
People-first
An extrovert who enjoys meeting new people and working in a team — I grow through it, and I'll step up to help lead when it's needed.
Self-awareness
I have a clear sense of my strengths and my gaps, and I stay humble about the gaps — there's always a next thing to get better at.
02Dispatch

Jan 2026 — Jun 2026

Armakuni

Software Engineer Intern

Python department. Built production-style backends and asynchronous processing pipelines, shipping from design to deployment in a fast-paced team.

  1. 01

    Built scalable services with Python, FastAPI, PostgreSQL, Redis, Docker, and AWS, focused on performance, reliability, and maintainability.

  2. 02

    Delivered projects spanning task management, movie-ticket booking, AI text-to-SQL, newsletter automation (FlashRead), and multi-agent AI systems (Boomerang).

  3. 03

    Gained hands-on experience with distributed systems, cloud-native architecture, and LLM application patterns.

Stack —Python /FastAPI /PostgreSQL /Redis /Docker /AWS

03Feature Well

Selected work, in priority order.

Backends have no screenshots, so these lead with what was built and, where there's a number worth showing, the number.

01

Distributed Systems

CineBook

Concurrency-safe movie-ticket booking backend.

A concurrency-safe movie-ticket booking backend that stops two people from grabbing the same seat, using Redis Lua distributed locks with PostgreSQL as the source of truth and Elasticsearch for search.

41msavg latency (92ms p95)0double-bookings @ 220 concurrent82%test coverage

  • Redis Lua-based distributed locking with all-or-nothing seat reservation and automatic TTL expiry.

  • Passwordless auth via email OTP + Google OAuth 2.0, 3-role RBAC, and encrypted QR tickets.

  • Validated with Locust: 60 concurrent users, 2,124 requests over 90 seconds, 99.3% rate-limit enforcement, zero race conditions.

FastAPI /PostgreSQL /Redis /Elasticsearch /SQLAlchemy /Locust /Docker

Source
02

Async Backend

FlowBoard

Async task platform with scheduled background work.

An async task-management API with OTP auth, Celery-scheduled reminders, S3 file uploads, and Redis rate limiting — asynchronous end to end.

53test suite (~75% cov)6containerized services60sreminder dispatch loop

  • Fully async FastAPI + SQLAlchemy 2.0 with Alembic migrations and a clean repository/service layering.

  • Celery worker + beat for scheduled reminders; Redis for caching and IP-based rate limiting.

  • AWS S3 presigned-URL uploads, request logging with sensitive-data masking, role-based admin panel.

FastAPI /PostgreSQL /Redis /Celery /AWS S3 /Docker /Alembic

Source
03

AI / LLM

Prompt2SQL

Natural language to safe, executable SQL.

Connects to any PostgreSQL database and turns plain-English questions into executable SQL, using dynamic schema introspection so the same instance adapts to whatever database it's pointed at. A strict SELECT-only safety layer blocks write operations before anything reaches the database.

120BGPT-OSS via GroqSELECTonly — write ops blockedRedisconversation memory

  • Dynamic schema introspection lets one instance adapt to any connected PostgreSQL database at runtime.

  • SELECT-only safety guard rejects INSERT, UPDATE, DELETE, DROP, and ALTER before execution.

  • OAuth + OTP authentication, with Redis holding per-session conversation memory; optional Streamlit frontend.

FastAPI /PostgreSQL /Redis /Groq (GPT-OSS-120B) /Streamlit

Source
Every repository on GitHub
04By the Numbers

Language mix

@Jaymin4724
Share of code by language across pinned repositories
LanguageShare
Python
58%
JavaScript
32%
TypeScript
6%
Other
4%

Focused on backend systems, AI/LLM applications, and clean architecture.

Pinned repositories

05Index

The full working vocabulary.

Grouped by where each one lives in a system — from the language up to the cloud it runs on.

Languages

05
  • Python
  • C++
  • JavaScript
  • TypeScript
  • SQL

Backend

09
  • FastAPI
  • Node.js
  • Express.js
  • SQLAlchemy
  • REST APIs
  • AsyncIO
  • Celery
  • JWT
  • OAuth2

Databases

04
  • PostgreSQL
  • Redis
  • SQLite
  • MongoDB

AI / LLM

08
  • Prompt Engineering
  • RAG
  • LangChain
  • LangGraph
  • Groq
  • Gemini
  • Embeddings
  • ChromaDB

Cloud & DevOps

12
  • Docker
  • AWS EC2
  • RDS
  • S3
  • Lambda
  • IAM
  • ECR
  • ECS
  • CloudFront
  • VPC
  • SNS
  • SQS

Testing

04
  • Pytest
  • TDD
  • Mocking
  • Locust

Tools

05
  • Git
  • GitHub
  • Linux
  • Postman
  • Alembic
In developmentCurrently building

ProgressOne

A new product I'm building right now. Full details coming soon.

On the pressSetting…

Next in Jaymin Dave

06Record

Wins, degrees, and the road here.

  • MSTC DAIICT

    1st Prize — Winter of Code 7.0

    Won the Node.js category (60 participants, 350+ overall) by building aDApt, a collaborative student platform with 15+ REST APIs and real-time chat.

  • Top 3 of 1,500+

    3rd Place — HackOut'24

    Placed 3rd among 1,500+ participants by shipping a full-stack solution in a 36-hour hackathon.

  • 2024 — 2026

    M.Sc. Information Technology

    DA-IICT, Gandhinagar. CPI 8.59.

  • 2021 — 2024

    Bachelor of Computer Application

    St. Xavier's College, Ahmedabad. CGPA 8.18.

  • 12th

    Higher Secondary — 12th (GHSEB)

    M. B. Patel Gyanjyot. 99.19 percentile.

  • 10th

    Secondary — 10th (GSEB)

    M. B. Patel Gyanjyot. 97.88 percentile.

07Correspondence

Let's build something reliable.

I'm open to backend and full-stack roles. The fastest way to reach me is email — copy it below, or find me anywhere on this list.

Open mail

Elsewhere