I Build AI Systems That Actually Ship.
About
Krish

My Approach
I treat every ML system like infrastructure, not a science experiment. That means drift detection, failure fallbacks, cost guardrails, and clean documentation — so your team owns it long after our engagement ends.
Where I Come From
Mathematics and software engineering background. I've shipped computer vision pipelines, on-device OCR apps, multi-agent systems, and enterprise RAG architectures — both solo and embedded with product teams.
What You Actually Get
Weekly demos, clear communication, and no disappearing after deployment. You get a documented, monitored, maintainable system — with observability baked in from day one. Reliability isn't a premium add-on. It's the baseline.
Professional
Experience
Freelance ML & Software Engineer
Ship production-ready ML systems for startups and engineering teams — RAG pipelines, autonomous agents, and MLOps infrastructure. Every project includes monitoring, documentation, and full handover.
AI & Data Engineering Trainee (Internship)
Completed a 78-hour GTU Skill-Based Training internship in collaboration with Microsoft. Achieved perfect scores (30/30) across 5 Microsoft-aligned assessments: AZ-900, AZ-104, AI-900, AI-103, and GH-300. Gained practical expertise in Azure Cloud Services, Azure AI, Generative AI, and AI-assisted development using GitHub Copilot.
AI & Quantum Technologies Trainee (Internship)
Completed a 4-week virtual internship in Emerging Technologies via Edunet Foundation, AICTE, and IBM SkillsBuild. Trained on Agentic AI, Cyber Security, and Quantum Computing. Architected and deployed an AI + Cloud capstone project. Earned IBM-recognized digital credentials and SkillsBuild badges.
Machine Learning Engineer Intern
Built production-ready salary prediction and fraud detection models using scikit-learn pipelines — from feature engineering through evaluation and deployment.
Flutter Developer Intern
Built the ML integration layer for SnapText, an on-device OCR app — enabling real-time text extraction without cloud dependency.
PHP Developer Intern
Built full-stack web modules using PHP, MySQL, and JavaScript — foundational engineering experience that shaped my approach to production systems.
Core
Services
Enterprise RAG
Your team shouldn't spend hours hunting through scattered documents and getting wrong answers. I build RAG systems with hybrid search, reranking, and evaluation built in — so the right answer surfaces in seconds, not hours.
- Self-Querying & Hybrid Search
- Advanced Chunking Strategies
- RAGAS Evaluation & Observability
- Multi-Vector Indexing
Agentic AI
From internal research assistants to customer-facing automation — I build multi-agent systems that handle complex workflows while keeping humans in the loop where it matters. Monitored, production-deployed, with proper safeguards.
- LangGraph/CrewAI Frameworks
- Function Calling & Tool Use
- Memory & State Management
- HIL (Human-in-the-loop) Design
MLOps Architect
Models that only work in notebooks don't generate revenue. I build the full ML lifecycle — reproducible pipelines, experiment tracking, drift detection, and CI/CD for ML — so your models stay accurate long after go-live.
- CI/CD for Machine Learning
- ZenML/MLflow Infrastructure
- Model Monitoring & Drift Detection
- Kubernetes/Cloud Orchestration
Selected
Case Studies
Production AI systems, persistent agent architectures, and end-to-end MLOps engines built for scale.
Technical
Resources
Curated technical documentation, course notes, and verified credentials across AI, systems architecture, and engineering.
Trusted by
Industry Leads
"Krish delivered a production RAG system in under 3 weeks. The handover documentation was thorough enough that our junior engineer could maintain it independently."
"His MLOps setup saved us from what would have been weeks of manual pipeline work. Everything was tracked, versioned, and reproducible from day one."
Let's Build
Something
That Works
Got a project in mind? Tell me what you're working on and I'll get back within 24 hours with honest thoughts on fit, scope, and timeline.















