Projects

Production systems and personal R&D — enterprise document intelligence, offline AI products, fine-tuned models, and more.

Offline AI Desktop Product

Production · Internal

Fully offline desktop app automating trial-balance migration and financial-statement reconciliation for an air-gapped enterprise audit client — deterministic-first matching with an embedded local LLM as strict last resort.

Pythonllama.cpppywebviewpdfplumberPyInstallerGitHub Actions

Fine-tuning Gemma 4 31B for On-Prem Document Analysis

Personal R&D

End-to-end fine-tuning pipeline for structured document analysis, trained on 100% synthetic data — QLoRA on a rented A100, GPTQ quantization, vLLM serving, and formal evaluation against gold-standard labels.

transformersPEFT / TRLQLoRAGPTQvLLMRunPodClaude

Document Intelligence Platform

Production · Internal

Enterprise platform that classifies, extracts, and analyzes financial documents using LLM-powered workflows with async FastAPI and AWS Bedrock.

PythonFastAPIAWS BedrockClaudePostgreSQLS3TextractDocker

Dynamic Report Generator

Production · Internal

Backend service that dynamically generates professional Excel reports from JSON templates and live API data, powering 40+ document workflows.

PythonFastAPIOpenPyXLPandasAWS S3Docker

Multi-Tenant Analytics Platform

Production · Internal

Enterprise backend powering data analytics, async workflow orchestration, and multi-database querying with database-per-tenant isolation and Keycloak OAuth2 security.

PythonFastAPIPostgreSQLKeycloakMulti-TenancyHatchetDockerOpenTelemetry

E-Commerce Data Ingestion Pipeline

Production · Internal

High-throughput data pipeline processing 15M+ product SKUs daily from hundreds of retailers via FTP, SFTP, and direct API feeds into a unified commerce data platform.

PythonGoFastAPIPostgreSQLData PipelinesETL

Distributed Web Crawler

Production · Internal

Large-scale distributed crawler processing 10M pages per day across 300+ e-commerce websites using a leader-follower architecture with 400 nodes spanning AWS, GCP, and third-party proxies.

GoPerlRabbitMQPostgreSQLRedisAWSGCPDistributed Systems