I architect production-ready backend systems, scrape multi-source data feeds, and deploy pragmatic LLM workflows. Delivered 12+ paid projects for international clients since 2020.
I design and implement custom, reliable backend and AI solutions that automate workflows, structure data, and fuel product growth.
Custom autonomous agents and intelligent search engines that streamline operations and eliminate repetitive human workflows.
Goal: Automate customer support, speed up document analysis, and construct intelligent RAG systems.
Robust data collection engines, automated web scrapers, and ETL pipelines designed to turn raw web noise into structured business databases.
Goal: Harvest data at scale, synchronize multi-source feeds, and automate business intelligence.
Secure, low-latency API architectures designed to support web and mobile products without scaling bottlenecks.
Goal: Establish rock-solid backend infrastructure with high availability and comprehensive telemetry.
Guidance on technology selection, database schema mapping, system scaling paths, and assessing AI project feasibility.
Goal: Verify project scope and prevent costly re-architectures before writing code.
Explore how backend and AI engineering solved concrete operational bottlenecks and delivered clear ROI for startups.
85%
Latency Reduction
92%
Search Accuracy
$12k
Monthly Notional Savings
The client had a slow internal search utility looking up thousands of complex engineering documents, guidelines, principles, etc. The legacy system suffered from high retrieval latency (averaging 45s) and frequent LLM hallucinations at scale.
Architected a low-latency hybrid search retrieval-augmented generation (RAG) pipeline. Designed custom indexing protocols using FastAPI, Open Search Instance, and custom semantic reranking models to filter out noise. Engineered a stateful response engine with strict guardrails to prevent hallucinations.
Reduced search and retrieval latency down to 3 seconds. Scaled seamlessly to 1,200+ monthly active users (MAU) while cutting manual overhead, saving $12,000/month in operations cost.
68%
Support Automated
88%
User Retention
99.5%
Call Uptime
The client faced scaling constraints due to overwhelming support volume, resulting in delayed ticketing, lost orders, and customer churn. Manual CRM entries consumed several hours of staff time daily.
Built a custom, CRM-integrated conversational AI engine. Developed a robust dialog state machine using Python, RASA NLU, and FastAPI to handle ordering inquiries and call routing. Integrated real-time webhooks with the client's CRM to auto-update transaction records and support transcripts.
Successfully deflected and automated 68% of inbound support inquiries. Boosted user retention to 88% while freeing up the core team from repetitive administration.
I write detailed walkthroughs of systems architecture, operational gotchas, and performance optimization lessons learned from client deployments.
Exploring how AI can automate and optimize administrative workflows through intelligent orchestration and generation.
A deep dive into common pitfalls and production challenges when deploying RAG (Retrieval Augmented Generation) microservices.
A deep dive into common pitfalls and production challenges when deploying RAG (Retrieval Augmented Generation) microservices.
Professional journey and key achievements
Technologies and expertise I work with
Academic journey and achievements
The Northcap University • 2022 - 2026
Sri Sri Academy • 2020 - 2022
Sri Sri Academy • 2020
Professional certifications and achievements
Learned NoSQL, Mongo Workspaces and database connectivities to applications.
View CertificateUnderstanding of cloud computing concepts and Azure services. Deployment of ML and AUTOML Apps onto the Azure and Databricks.
View CertificatePracticed Cloud Data Engineering and validated knowledge on ETL pipelines.
View CertificateLet's map out your systems architecture, assess the feasibility of your AI workflows, or scope out a high-performance database plan.
Let's connect and discuss opportunities