Software Engineer · AI Products
Building AI-powered products & the systems behind them.
I design and ship intelligent software end-to-end — from LLM-backed features and retrieval pipelines to the cloud-native backends that keep them reliable. I work closely with product and engineering teams to iterate quickly and deliver at the core-platform level.
Duration 8:50:13Featured video · 15 Mar 2026
MySQL Full Course 2026 | Zero to Hero (9 Hours) (opens on YouTube)whoami
I build AI-powered products and the reliable systems behind them.
I'm a software engineer with 4+ years of experience building AI products, cloud-native backend systems and data platforms. At EviView I work with product and engineering teams to turn ideas into features and dependable production systems. I enjoy connecting thoughtful system design with practical product delivery.
4+
Years building software
M.Sc.
Advanced Computer Science
6
Teams & clients shipped for
git log --author=priyanshu
Developing and improving EviView's in-house AI-powered products, working closely with product and engineering to build, iterate on and ship features for the core platform.
Tata Consultancy Services
Built and optimised high-throughput serverless backends for a leading North American airline. Tuned SQL using explain plans and added Redis caching to cut query latency and Lambda cold starts. Secured APIs with API Gateway, Cognito and custom JWT authorisers, ran CI/CD through CloudFormation nested stacks, and built event-driven pipelines on Kafka monitored with CloudWatch and Kibana.
Triluxo Technologies
Built an intent-driven conversational agent for hotel service automation using Google Dialogflow and Node.js, including the NLP routing logic and webhook integrations that turn free-text requests into structured database operations.
Gallant Fusion Tech
Designed the MySQL schema and data models for a multiplayer gaming application and built its REST API in Node.js and Express, with Passport-JWT authentication.
ls ~/projects
Featured project
Code-aware RAG engine for GitHub repositories
An agentic AI system that ingests GitHub repositories and lets you ask questions about them in natural language. A custom RAG pipeline uses AST parsing to chunk code at function and class boundaries, so retrieved context stays semantically whole. OpenAI embeddings stored in Supabase (pgvector) power similarity search, and answers are grounded in the retrieved code to keep the model from making things up.
Inventory & dispatch system
A full inventory and dispatch system tracking incoming and outgoing steel products, with an analytics dashboard over historical sales data to surface supply-chain bottlenecks.
Restaurant, United Kingdom
A contactless QR menu connected to a central POS dashboard, plus a data pipeline that aggregates daily orders into revenue and demand analytics.
Private client deployment — no public link
B2B presence for a steel trader
A fast, responsive corporate site for an industrial steel-trading business, built with attention to SEO and load time.
cat stack.yaml
LLM-backed features, retrieval and agentic workflows.
High-throughput, low-latency services.
Automated, zero-downtime delivery on AWS.
Reliable pipelines and analytical workloads.
Caching tiers and well-tuned relational databases.
cat education.log
Swansea University
Focused on cloud computing, artificial intelligence and distributed systems, with an emphasis on scalable software architecture.
Sagar Institute of Science and Technology
Foundations in data structures, algorithms, systems software design and database systems.
ping priyanshu
Whether you're building AI products, scaling backend systems, or just want to talk tech — my inbox is open.
