Prasath Rajasekaran

Full Stack Engineer

Full Stack Engineer with 6 years of experience in high growth fintech startups. Scaled products from 0 -> 1 as part of multiple founding teams to hundreds of thousands of users globally.

PythonRubyTypeScriptSQLBashFastAPIRuby on RailsReactGraphQLLangChain
Swipe to start

Skills & Certifications

Programming Languages & Frameworks

PythonRubyTypeScriptSQLBashFastAPIRuby on RailsReactGraphQL

AI & LLM Engineering

LangChainRAG PipelinesAgentic WorkflowsGemini ModelsVector SearchElasticsearch

Cloud & Infra

AWS (Lambda, S3, EC2, RDS, SAM, Eventbridge)DockerGCPCircleCI

Databases & Storage

PostgreSQLMySQLMongoDBDynamoDBRedisElasticsearch

Messaging & Monitoring

SQSCloudWatchDatadogPub/Sub

Tools

GitCursorClaude Code

Certifications

Solutions Architect Associate
AWS
Certified Scrum Product Owner
Scrum Alliance
1
AI Driven Customer Insights Analysis

AI Driven Customer Insights Analysis

AWS LambdaSQSLangChain

Event-driven Agentic pipeline analyzing 10k+ weekly tickets for root-cause insights.

Tap anywhere to flip

AI Driven Customer Insights Analysis

Overview

Architected an event-driven Agentic pipeline (AWS Lambda/SQS) analyzing 10k+ weekly tickets for root-cause insights. Implemented RAG-based compliance workflows to validate agent responses against policy docs, saving 20+ weekly audit hours.

Tech Stack

AWS LambdaSQSLangChainRAGDynamoDBGemini API

The Challenge

Our product and ops teams were manually analyzing over 50,000 Zendesk support interactions monthly. This process was slow, and we were missing real-time insights into critical customer pain points.

The Solution

I took the initiative to design and build an end-to-end, serverless AI pipeline. I used AWS Lambda and EventBridge to process the interactions in real-time, integrating with the Gemini API to automatically extract, classify, and analyze the sentiment of each customer issue. The data was then stored in DynamoDB.

⚡ The Impact

Built a real-time dashboard on top of this data that provided immediate visibility and drill-down capabilities. This empowered our product team to move from slow manual analysis to instant, data-driven decisions, which directly informed our product roadmap and feature prioritization.

Built a real-time dashboard on top of this data that provided immediate visibility and drill-down capabilities. This empowered our product team to move from slow manual analysis to instant, data-driven decisions, which directly informed our product roadmap and feature prioritization.

2
LinkBee - AI LinkedIn Agent

LinkBee - AI LinkedIn Agent

Chrome ExtensionGemini 2.5 FlashVite

Chrome Extension acting as your personal technical career coach and executive assistant for LinkedIn.

Tap anywhere to flip

LinkBee - AI LinkedIn Agent

Overview

LinkBee is a powerful Chrome Extension that acts as your personal technical career coach. It scans your messaging inbox, analyzes conversations using Google Gemini AI, and identifies high-value opportunities to follow up—ensuring you never drop the ball on a recruiter, peer, or lead.

Tech Stack

Chrome ExtensionGemini 2.5 FlashViteJavaScriptTailwind

The Challenge

Job seekers and professionals struggle to manage hundreds of LinkedIn conversations. High-value opportunities (recruiter messages, referral asks) often get lost in the noise, and "dead threads" are rarely revived effectively because manual tracking is impossible at scale.

The Solution

I built a Chrome Extension powered by Google Gemini 2.5 Flash. It utilizes a Service Worker to scan the inbox, auto-scrolls to find old chats, and applies a fuzzy-matching algorithm to locate specific people. The AI classifies chats into 7 strategic playbooks (e.g., "Recruiter Recovery", "Strategic Pivot") and drafts context-aware replies.

⚡ The Impact

Transforms a chaotic social inbox into a prioritized Sales Pipeline. Features "Smart Navigation" that handles LinkedIn's complex SPA architecture without reloading, automated confidence scoring, and context-aware draft suggestions.

Transforms a chaotic social inbox into a prioritized Sales Pipeline. Features "Smart Navigation" that handles LinkedIn's complex SPA architecture without reloading, automated confidence scoring, and context-aware draft suggestions.

3
High-Performance Payment Gateway

High-Performance Payment Gateway

PythonPostgreSQLRedis

High-performance payment gateway handling 200k+ users with row-level locking.

Tap anywhere to flip

High-Performance Payment Gateway

Overview

Engineered concurrency control for payment microservice using row-level locks, improving system reliability for 200k+ users. Reduced median, API latency by 250ms.

Tech Stack

PythonPostgreSQLRedisAWSRow-Level Locking

The Challenge

We had a critical customer-reported bug where payments were failing at a specific time daily. Initial investigation showed no code errors, but the failures were real, indicating a deep system-level conflict.

The Solution

I conducted a deep system analysis and discovered a race condition: a background installment update job was locking database rows at the exact same moment users were trying to pay. I implemented a robust Postgres Row-Level Lock strategy to act as a "traffic cop", ensuring user transactions always took precedence.

⚡ The Impact

This fix completely eliminated the "ghost" payment failures and stabilized the core financial transaction flow for over 200,000 users. It transformed a flaky experience into reliable infrastructure.

This fix completely eliminated the "ghost" payment failures and stabilized the core financial transaction flow for over 200,000 users. It transformed a flaky experience into reliable infrastructure.

4
Automated e-KYC

Automated e-KYC

PythonFastAPIGoogle Vision API

Automated identity verification system reducing onboarding time by 95%.

Tap anywhere to flip

Automated e-KYC

Overview

Built a lightweight, high-impact identity verification service to replace a manual admin process. The system automatically extracts, validates, and processes user ID documents.

Tech Stack

PythonFastAPIGoogle Vision APIOCR

The Challenge

New user onboarding was a manual bottleneck. Admins had to physically look at uploaded IDs and type data into the dashboard. This caused a 3-day wait time for users and high drop-off rates.

The Solution

Instead of over-engineering a custom ML model, I initiated a "scrappy" solution using FastAPI and the Google Vision OCR API. I built a wrapper service that accepted ID images, sanitized the OCR output, and auto-filled the admin review form.

⚡ The Impact

slashed user verification time from 3 days to near-instant. Automated 95% of all applications, leaving admins to handle only the edge cases.

slashed user verification time from 3 days to near-instant. Automated 95% of all applications, leaving admins to handle only the edge cases.

5
GitResume

GitResume

PostgreSQLGraphQLFastAPI

AI-driven service transforming Git commits into quantified resume bullets.

Tap anywhere to flip

GitResume

Overview

Built an AI-driven service that transforms Git commits into quantified resume bullets, reducing self-reporting effort by 80%. Operates in the $100–500M developer résumé tools (TAM), validated on 500+ commits.

Tech Stack

PostgreSQLGraphQLFastAPIReactLLM

The Challenge

Engineers struggle to remember and quantify their impact when writing resumes. Their actual work is buried in thousands of Git commits.

The Solution

I built an Agentic workflow that connects to GitHub, analyzes commit history using LLMs, and automatically generates "Resume Ready" bullet points using the STAR method.

⚡ The Impact

Validated on 500+ commits and reducing the time to write a technical resume by 80%.

Validated on 500+ commits and reducing the time to write a technical resume by 80%.

6
Reduced build/test time by 50% via parallelization.

Reduced build/test time by 50% via parallelization.

CircleCIDockerBash

Intelligent CI/CD pipeline optimizing deployment efficiency and developer feedback loops.

Tap anywhere to flip

Reduced build/test time by 50% via parallelization.

Overview

Architected a sophisticated CircleCI pipeline for the Core Platform API, focusing on reducing cycle time and "shifting left" on quality and security. The pipeline orchestrates testing, security scanning, and multi-environment deployments while keeping project management tools in perfect sync.

Tech Stack

CircleCIDockerBashGitLab APIClickUp APIRuby

The Challenge

The engineering team faced slow feedback loops due to sequential testing, manual ticket updates in ClickUp, and silent failures in code quality checks. Deployments were a bottleneck.

The Solution

I engineered a parallelized pipeline (CircleCI) that runs RSpec across 4 concurrent nodes. I wrote custom Bash scripts to parse Bundle Audit & Rubocop results and post them directly as comments on GitLab Merge Requests. I also built automation to move ClickUp tickets based on branch status.

⚡ The Impact

Reduced build/test time by 50% via parallelization. The "Bot Feedback" on MRs enforced zero-vulnerability standards without human intervention. Ticket automation saved ~2 hours of admin work per dev/week.

Reduced build/test time by 50% via parallelization. The "Bot Feedback" on MRs enforced zero-vulnerability standards without human intervention. Ticket automation saved ~2 hours of admin work per dev/week.

Want to know more?

Download my resume to see the full picture.

Download Resume

© 2026 Prasath Rajasekaran