~/krishna_dhakal
Building RAG Pipelines with AWS Bedrock
> June 10, 2025
A deep dive into architecting retrieval-augmented generation pipelines using AWS Bedrock, Qdrant vector store, and llama.cpp for production AI applications.
// read_more →Local LLM Development with llama.cpp and HuggingFace
> May 20, 2025
How to run and evaluate large language models entirely on-premise using llama.cpp and HuggingFace Transformers — benchmarking local inference against cloud-hosted counterparts on latency, accuracy, and retrieval relevance.
Building a Real-Time Video Assistant with RAG on AWS
> April 15, 2025
A walkthrough of architecting an end-to-end RAG pipeline that scrapes live news, indexes it into Qdrant, and autonomously joins Microsoft Teams meetings to deliver contextually relevant briefings — built entirely on AWS.
Building TextQL — an LLM-Powered Text-to-SQL Pipeline
> September 5, 2024
How we built TextQL, an open-source tool that converts natural language questions into structured SQL using Google Gemini and FastAPI — with prompt engineering patterns that make LLM output reliably parseable.
// projects
Multi-signature transaction platform with a Go backend and PostgreSQL, implementing cryptographically secure payment orchestration workflows.
Next.js educational platform for children, with dynamic content management and interactive learning modules.
LLM-driven text-to-SQL pipeline using Google Gemini and FastAPI; normalizes natural language queries into structured SQL via prompt-engineered LLM calls.
Automated retail asset generation using Tesseract OCR, Google Custom Search, and PIL — eliminating manual design work for weekly promotional materials.
Observability stack using Grafana Loki, Prometheus, and Kubernetes for centralized log monitoring and alerting at scale.
Zero-setup feedback widget for static websites — users embed a single script tag to collect structured feedback. Built with Supabase, React, Netlify Functions, and Cloudflare Turnstile.
// experience
$ March 2025 – Present
Software Engineer – AI/ML & Full-Stack @ Leapfrog TechnologyArchitected end-to-end RAG pipelines and AI workflow automation using AWS Bedrock, Qdrant, and llama.cpp. Built an AI-powered Video Assistant and a conversational chatbot for a Nursing Management Platform with agentic orchestration over REST APIs.
$ Jan 2024 – July 2024
Professional Services Consultant – AI & Search @ Upland SoftwareEngineered enterprise-scale semantic vector search over 2M+ SharePoint documents using Azure OpenAI embeddings and Azure AI Search. Delivered an OCR + OpenAI pipeline for an Australian client to extract and index structured data from unstructured PDFs.
$ Nov 2023 – April 2024
Software Engineer – Full-Stack @ Sherpa TechnologiesLed migration of a credit card dispute management platform from legacy infrastructure to GCP Cloud Functions using Node.js, TypeScript, and PHP. Drove cross-functional performance triage across client-facing web and mobile applications.
$ April 2021 – April 2022
Software Engineer – Blockchain & Web3 @ 0chain LLCRe-architected a React-based blockchain explorer and maintained two Go CLI tools (zwalletCLI, zboxCLI) for 0chain's decentralized storage protocol. Migrated the business website to a Netlify CMS + GatsbyJS stack and contributed to a TypeScript + WebAssembly storage migration.
$ April 2020 – Nov 2023
Full-Stack Developer @ FreelanceBuilt production-grade platforms including a real-time media app (AWS S3, Redis, WebSockets) and SEO-optimized 3D animation sites using Three.js. Automated client workflows with Pipedream and AssemblyAI, and upgraded a financial trading platform built on Angular and TradingView APIs.
// products
// under_construction
// testimonials
"John's work is fantastic! He delivered exactly what we needed."
- Jane Smith, CEO of XYZ Company
"Working with John was a pleasure. He's professional and skilled."
- Bob Johnson, Freelance Designer
// roadmap
- Learn advanced JavaScript frameworks
- Contribute to open-source projects
- Develop a mobile app