Full-Stack Engineer · Go · AI · Distributed Systems

I’m Ashish Yadav.
I Build Fast Systems for
Problems That Don’t Stay Small.

I’m a Full-Stack Engineer focused on Go, distributed systems, cloud infrastructure, and production AI. I’ve architected systems across 50+ product domains and shipped AI infrastructure for a 100K+ download app with sub-200ms backend latency.

View ResumeExplore My Work
Theme:
50+
Product Domains
100K+
App Downloads
<200ms
AI Backend Latency
GO•AWS•REACT•NEXT.JS•DISTRIBUTED SYSTEMS•AI
Career History

Production Experience & Impact

Architecting resilient serverless backends, high-throughput cloud infrastructure, and production AI systems at scale.

Currently Shipping at ASOasis Tech
AT

Software Developer Intern

Internship
ASOasis TechRemote
Aug 2025 — Present

Building production-grade backend, AI, cloud, and workflow automation systems across FinTech, SaaS, and AI products.

Key Performance Metrics
50+Product domains
100K+App downloads
<200msAI backend latency
6+Grounded AI features
~95%Repeat query reduction
Engineering Scope & Deliverables

Architected a serverless Go backend spanning 50+ product domains on DynamoDB, enforcing zero-Scan reads and 15+ shared rate limiters for scalable production workloads.

Shipped a real-time AI chat backend for a 100K+ download consumer app with sub-200ms latency, combining Gorse recommendations, Redis, MySQL, and a sigmoid-based matching model across 5 engagement signals.

Built 6+ grounded-generation AI features including adaptive mock interviews, resume tailoring, and promotion packets, enforcing server-verified citations and blocking unsupported AI output before persistence.

Delivered a multi-tenant workflow automation platform using EventBridge scheduling and 2–8 concurrent workers, alongside a BigQuery analytics backend that reduced repeat query volume by ~95% through 5-minute TTL caching.

Hardened production systems with SSRF protection, no-egress sandboxing, PII-minimized logging, and GDPR export/erasure workflows, while tracing and fixing a live rate-limit bypass using signed-header trust verification.

Technologies & Architecture
GoAWS LambdaDynamoDBS3EventBridgeSQSBigQueryRedisMySQLGorseReactNext.jsAI / RAG
Selected Work

Systems & Production Projects

Distributed schedulers, multimodal notification routers, content-addressed storage, and custom RAG engines built for scale.

ReachInbox

Featured

Distributed Email Job Scheduler & Analytics Platform

Full-stack distributed email infrastructure for scheduling, throttling, dispatching, monitoring, and recovering high-volume outreach campaigns.

315msBatch scheduling
5K leadsCampaign size
O(1)Rate limiter
Key Implementations
Built a zero-cron distributed scheduler using BullMQ delayed queues backed by Redis ZSETs, with PostgreSQL as the persistent job state machine.
Designed atomic O(1) Redis Lua token-bucket rate limiting with adaptive cooldowns for SMTP 429/503 failures.
Optimized 5,000-lead campaign ingestion using PostgreSQL UNNEST and BullMQ addBulk(), reducing thousands of network operations to 2 calls.
Next.jsReactTypeScriptNode.jsExpressPostgreSQLRedisBullMQ

WhatsApp Signal Router

Featured

AI-Powered Multimodal Notification Router

Multimodal AI system that analyzes incoming WhatsApp messages and intelligently decides whether to notify, digest, or mute them.

86.7%Action accuracy
100%Mute precision
3Routing modes
Key Implementations
Built an end-to-end multimodal pipeline handling text, images, and voice notes using Gemini Vision OCR and local Faster-Whisper transcription.
Combined deterministic safety rules with LLM reasoning for messages involving OTPs, money, links, urgency, and unknown senders.
Implemented a confidence-weighted safety fallback that applies conservative routing when the model is uncertain.
PythonGeminiFaster-WhisperLLMOCRASRRule Engine

Mini-Dropbox

Distributed File Storage System

Go-based distributed file-storage system with master-node coordination, content-addressed storage, deduplication, and cross-node retrieval.

2,000 req/sThroughput
4×Performance
43%Memory reduction
−55%Retrieval time
Key Implementations
Engineered distributed file storage in Go with master-node coordination and cross-node retrieval.
Implemented content-addressed storage and deduplication using MD5-based file identity.
Designed concurrent-safe storage operations using RWMutex across storage nodes under load.
GoREST APIDistributed SystemsConcurrencyCachingsync.Pool

ASHOS

AI-Powered CLI Personal OS

Event-driven Go CLI combining RAG, hybrid vector search, modular infrastructure, and AI-powered productivity workflows.

4+Data sources
4AI commands
6+Concurrent modules
Key Implementations
Built an event-driven CLI OS in Go using dependency injection and interface-driven storage.
Implemented a RAG engine from scratch using OpenAI embeddings, SQLite-vec, custom cosine similarity, and GPT synthesis.
Integrated 4+ data sources through an asynchronous ingestion pipeline.
GoOpenAIRAGSQLiteSQLite-vecVector Search
Engineering Profile & Telemetry
Go 1.24 Microservice · TTL Cache 30m

About Me & GitHub Ecosystem

Live systems telemetry, verified GitHub GraphQL contributions, repository architecture & algorithmic metrics.

Public Repositories
59

Distributed, Go & AI codebases

Yearly Contributions
344

Verified GraphQL sync

Commit Velocity
+160%

Surge vs. prior 30-day window

Peak Streak
47 days

Peak activity: Mon

GitHub Verified
@Ashish5180
Ashish Yadav

Ashish Yadav

@Ashish5180

Full-Stack Developer | Merging MERN with Go lang , Flutter & Next.js | Where Product Meets Performance

8 Followers·10 Following
India · Remote
K.R Mangalam University

Awards & Milestones

6 Earned
50 Days
50 DaysStreak Badge
100 Days
100 DaysActive Streak
300+ Solved
300+ SolvedMilestone
Contest Knight
Contest KnightProblem Solver
CodeChef Star
CodeChef StarRated Division
Contest Star
Contest StarCompetitive

GitHub Contribution Calendar

52-Week Flexible Horizontal Stream
Submissions: 344Streak: 47 days
↔ Scroll horizontally to explore history
MonWedFri
Hover on any cell for detailed commit count
LessMore

Algorithmic Distribution

DSA difficulty & competitive programming breakdown

131DSA
Easy33
Medium75
Hard23
172Contests
CodeChef89
Codeforces83

Topic Proficiency

Problems Solved
Arrays73
Math42
String37
HashMap and Set23
Dynamic Programming22
Sorting20

Live Go Microservice Benchmarker

:8080 Active

Real-time TCP network latency & concurrent goroutine stress-testing.

Client Browser
↔ TCP Round-Trip Pipeline
🏓
Go Server (:8080)
➜PONG! Ready for TCP SYN-ACK handshake rally. Click to test!
Rally #0
Runtime: Go 1.24·Active Goroutines: 3
99.999% SLA Guaranteed
Global Reach & Collaboration

Remote Connectivity

Distributed asynchronous collaboration across timezones. Available for engineering teams worldwide.

world map
Currently Open for Opportunities

Let’s Build High-Impact Systems Together

Whether you need a scalable Go/AWS backend, high-throughput distributed queue architecture, or grounded production AI pipelines — I’m ready to contribute immediately.

Timezone FlexibilityIST (UTC+5:30) · Overlaps US & EU
Work PreferencesRemote · SDE / FDE / Backend
contact@ashish.dev
Send Email

Connect Directly

Explore coding history, professional experience, and profiles.

LinkedInProfessional Network
GitHubRepositories & Code
Codolio Profile300+ Solved · 47-Day Streak
Designed & Engineered by Ashish Yadav · 2026