TejashwarReddyKatika
Building agentic systems, cloud-native orchestration, and the interfaces between them.
Currently
Building Artha · Gr8Saver · Karmada
Building
- →
Artha — agentic finance framework
Java 21 · Claude · arXiv
- →
Gr8Saver — AI deal aggregator
Next.js · DeepSeek · live
- →
Karmada — CNCF contributions
Go · multi-cloud
Learning
- →
Evaluation harnesses for agents
eval sets, traces, scoring
- →
Distributed scheduling internals
kube-scheduler, Karmada
- →
Vector DBs + ontology retrieval
pgvector, hybrid search
Thinking About
- →
Where agents replace microservices
architecture
- →
Ontologies as context for LLMs
typed tools beat JSON blobs
- →
Orchestration for AI clouds
infrastructure
I build systems that reason, scale, and adapt.
I'm Tejashwar — a software engineer working at the intersection of agentic AI, cloud-native orchestration, and the interfaces that make them usable.
Right now I'm researching Artha— a domain-ontology-driven agentic framework where a Claude agent reasons over typed financial objects instead of raw SQL rows. I'm also shipping Gr8Saver, a live AI deal aggregator with nine scrapers, DeepSeek V3 authenticity scoring, and a natural-language search agent — deployed and serving real traffic.
In parallel, I contribute to Karmada— the CNCF multi-cloud Kubernetes orchestrator — because the infrastructure problem of the next decade is running AI workloads across cloud boundaries, and that's where I want to be useful.
Current Role
Software Engineer
Automate IT Inc
Education
MS Computer Science
Univ. of North Texas
Live Product
Gr8Saver
gr8saver.com
Research
Artha — arXiv
agentic finance AI
The stack, without the buzzwords.
Depth marked with dots · hover a row for context.
Intelligence
Agentic AI Systems
— Artha's 15 typed tools, eval harness
LLM Integration
— few-shot, structured output
ML Engineering
— from pipeline to inference
Systems
Distributed Orchestration
— multi-cloud control planes
Cloud Architecture
— AWS SAA · CCP · active
Backend Engineering
— contract-first services
Data Platforms
— relational · cache · NoSQL
Interface & Delivery
Frontend
— app-router, a11y, motion
DevOps & Observability
— ship fast, watch closely
★ AWS Solutions Architect · AWS Cloud Practitioner · active CNCF contributor
Where I've worked
Software Engineer
Automate IT Inc
Sep 2024 — Present
Dallas, TX
- –Design, develop, and validate cloud edge software modules in Java and Python — end-to-end from design and coding through unit testing, debugging, and integration into production systems.
- –Translate user needs and software requirements into clear interface specifications and architectural documentation, then implement modules to match.
- –Develop and maintain REST APIs and GraphQL interfaces with well-defined contracts; perform systematic code review and regression testing.
- –Triage and fix defects precisely to maintain code quality and keep delivery on schedule.
Assistant Engineer
University of North Texas
Oct 2022 — May 2024
Denton, TX
- –Designed and implemented Java and Python software modules for internal platforms, applying OOP design patterns, data structures, and algorithmic precision.
- –Developed REST APIs with clearly defined input/output contracts and managed PostgreSQL/MySQL schema design.
- –Performed regression testing and code review across the team codebase — triaging production defects using IDE debug tools and Linux diagnostics.
- –Used Git with structured branching and pull-request workflows to support collaborative multi-engineer development.
Things I've built
drag to exploreIntelligence is
the new infrastructure.
Software engineering is undergoing its most profound shift since the cloud. Here's how I think about building for what comes next.
The Stack Is Collapsing Upward
AI isn't a feature you add to a system — it's becoming the system. Models now handle logic that once required entire microservice layers. Vector databases are replacing query engines. Agents are replacing workflows. The engineer's value is shifting from writing code to designing the architecture that generates, validates, and orchestrates it. Understanding every layer of the stack is no longer optional — it's the moat.
Scalability Now Means Adaptability
Traditional scalability handles load. The new scalability handles change — new models, drifting data distributions, swapped APIs, and user behaviours nobody predicted. Systems built for adaptability absorb these shifts without rearchitecting. That requires loose coupling, contract-first API design, and infrastructure capable of swapping AI providers the way you swap a database driver. Rigidity is the new single point of failure.
Orchestration Is the AI Frontier
As AI workloads grow, so does pressure on multi-cloud, multi-cluster infrastructure. Running inference across regions, coordinating GPU fleets, managing training pipelines at scale — this is the orchestration problem of the next decade. It's precisely why I contribute to Karmada. The engineers who understand distributed orchestration today will be the ones architecting the AI clouds of tomorrow.
Engineering Philosophy
“I build at the intersection of cloud infrastructure, intelligent systems, and the interfaces that make them usable — not because it's trendy, but because that's where the hardest problems live.”
— Tejashwar Reddy Katika
Academic & credentialed background
Degree
Master of Science — Computer Science
University of North Texas
Aug 2022 — May 2024 · Denton, TX
Certification
AWS Certified Solutions Architect
Amazon Web Services
Associate · Dec 2024 — Dec 2027
Certification
AWS Certified Cloud Practitioner
Amazon Web Services
Oct 2024 — Oct 2027
Let's build something together.
I'm open to new opportunities, collaborations, and interesting conversations. Whether you have a project in mind or just want to connect — reach out.
Say hellotejashwar1029@gmail.com