JOSH
AMBATI
Building production-scale AI systems, LLM orchestration, and full-stack products that create real business outcomes.
01—About
I engineer AI systems that ship. From multi-model inference to full-stack products — I build what others are still planning.
I grew up in India, got my engineering degree from LPU in Punjab, and started my career building enterprise software at CDK Global. That's where I learned the fundamentals — distributed systems, identity management, shipping production code that thousands of people depend on.
I moved to Chicago in 2023 to pursue my Master's at Roosevelt University, and somewhere along the way I realized I didn't just want to write code for other people's ideas — I wanted to build my own. That's when Inkrant started.
Inkrant is an AI product company I built from the ground up. The flagship product, Schools by Inkrant, is an AI-native operating system for schools — built with intelligent model routing, context-aware AI, and a level of engineering depth that most ed-tech doesn't even attempt.
Today I work as a full-stack developer at Peterson Technology Partners in Chicago, building AI-powered platforms during the day and scaling Inkrant in every other hour. I think in systems, I ship fast, and I believe the best software comes from people who actually care about the problem.
02—Work & Education
Where I've been
Peterson Technology Partners
Park Ridge, IL
Sep 2025 — Present
Full-time
Full Stack AI Developer
Engineered AI-powered screening system with Rebecca AI voice agent, reducing manual screening by 80%. Built enterprise ATS with React 18 handling 10+ modules serving 100+ recruiters. Designed real-time analytics dashboards, granular RBAC with caching, and integrated AWS Cognito across 15+ backend endpoints. Implemented optimistic UI patterns reducing perceived latency by 60%.
Roosevelt University
Chicago, IL
Aug 2023 — May 2025
Education
M.S. Computer Science
Master's degree with focus on AI systems, distributed computing, and software engineering. Applied research in LLM orchestration and memory-augmented generation to build Inkrant.
CDK Global
Austin, TX / Hyderabad, India
Aug 2021 — Jul 2023
Full-time
Associate Software Engineer
Developed automated customer migration system handling 100+ users in under 10 minutes, eliminating manual intervention. Worked with CI/CD pipelines, unit testing, and integration testing across distributed services. Part of the Identity and Access Management team securing authentication for thousands of internal users.
Lovely Professional University
Punjab, India
2018 — 2022
Education
B.Tech Computer Science & Engineering
Bachelor of Technology in Computer Science and Engineering. Head member of Student Organization - Otaku, leading events and activities. Earned certifications in Azure, MongoDB, and HackerRank problem solving.
03—What I Built
My product
inkrant.com
Inkrant
Software that moves industries.
Inkrant builds vertical AI products for industries that generic SaaS has overlooked. I built this company and its engineering from scratch. Schools by Inkrant is the flagship — an AI-first operating system for schools covering admissions, academics, attendance, payments, and communication.
Schools by Inkrant
Dashboard Preview
Students
1,284
Present
94%
Fees due
₹2.1L
Notices
3
Intelligent model routing
Dynamic AI routing that optimizes for cost and performance
Context-aware AI
Systems that learn and improve with every interaction
Vendor-independent infra
Flexible architecture that isn't locked to any single provider
15+ AI-powered features
Prediction, analytics, and automated decision systems
Other projects
AI Interview Platform
Production-scale AI screening system integrating Rebecca AI voice agent with intelligent retry logic and scoring algorithms. Processes thousands of candidates with automated evaluation pipelines.
Enterprise ATS
Full-featured Applicant Tracking System with 10+ interconnected modules — dashboard analytics, candidate management, RBAC, real-time streaming, and multi-tenant isolation.
04—How I Think
Engineering principles
These aren't things I read somewhere — they're lessons from shipping real products, debugging production at 2 AM, and building systems that thousands of people rely on every day.
01
“The model is 10% of the work.”
Context management, error handling, fallback logic, and user experience are the other 90%. Systems thinking > model selection.
02
“Ship fast, fix fast.”
Waiting for perfection is the biggest risk. Build for real users, get feedback from reality, iterate relentlessly.
03
“Architecture outlasts code.”
Good separation of concerns scales from a single function to an entire system. The decisions you make early compound forever.
04
“Own it end to end.”
The best work happens when one person owns the feature from database to pixel. No handoffs, no waiting — just shipping.
05—Interests
Roles I'm drawn to
I bring 3+ years of shipping production software, hands-on AI engineering from model routing to full-stack deployment, and the rare ability to own a product end-to-end — from database to pixel. I don't just write code, I build systems that drive real outcomes. Hover to see why I'm a fit.
Full Stack Developer
3+ years building end-to-end products with React, Spring Boot, Node.js, and cloud infra. Built Inkrant's entire stack from scratch.
AI Engineer
Shipped 15+ AI features in production — multi-model routing, LLM orchestration, prompt pipelines, and inference optimization at Inkrant and PTP.
AI Systems Engineer
Designed provider-agnostic inference architecture, memory-augmented generation, and dynamic model routing that cut costs by 70%.
Software Engineer
Strong fundamentals in distributed systems, microservices, OAuth2, CI/CD — built and shipped production code at CDK Global and PTP.
MLOps Engineer
Built model serving pipelines, monitoring dashboards with Prometheus/Grafana, and containerized deployments with Docker and Kubernetes.
Backend Engineer
Deep experience in Java/Spring Boot, Python, Node.js, PostgreSQL, MongoDB, Redis — plus IAM systems securing thousands of users.
06—Writing
Blog
Deep dives into AI architecture, engineering decisions, and lessons from building products in production. Click to read.
07—FAQ
Common questions
Things recruiters and hiring managers usually ask. Saved you an email.
React, Next.js, TypeScript, Python, Java, Spring Boot, Node.js on the application side. For AI — LLM orchestration, multi-model routing, prompt engineering, and inference optimization. Infrastructure includes AWS, Docker, Kubernetes, PostgreSQL, MongoDB, and Redis.
I'm currently based in Chicago, IL and open to opportunities here as well as remote roles. For the right opportunity, I'm absolutely open to relocation.
Full Stack Developer, AI Engineer, Software Engineer, MLOps, or Backend Engineer roles where I can build production systems and have real ownership over what I ship. I thrive in fast-moving teams that care about quality.
Inkrant is an AI product company I built from scratch. The flagship product — Schools by Inkrant — is an AI-native operating system for schools, handling admissions, academics, attendance, payments, and communication with 15+ AI-powered features.
Yes. I've built and shipped multi-model inference systems, dynamic model routing (LLaMA, GPT-4o, Gemini Flash, Mixtral), memory-augmented generation, voice AI agents, and prompt pipelines — all running in production serving real users.
I have a valid work authorization in the United States. Happy to discuss specifics during the interview process.
I can be flexible with start dates. Depending on the role, I can start within 2-4 weeks.
I've built an entire product company from zero — that means I've done everything from database design to deployment, from AI architecture to user-facing features. I don't just write code, I build systems. I think like an owner, ship fast, and care deeply about the end result.
08 — Contact
Let's build
something great.
Have a role, a project, or just want to talk AI and engineering? Always open.