AI SYSTEMS · BACKEND · PRODUCTS
From model capabilities
to reliable systems.
I'm Arjun, a software engineer focused on AI applications, backend architecture, and independent products. I document what it takes to move from prototype through deployment to continuous iteration.
✓ response streamed · trace saved
SELECTED WORK
What I'm building
ALL REPOSITORIES ↗MediaForge
An AI image platform for ecommerce content workflows, connecting multimodal retrieval, agents, quality evaluation, and batch processing.
Python · Multimodal RAG · Agents
Finance Tweet Analyzer
A financial tweet analysis platform that uses a multi-agent pipeline for retrieval and analysis, with results streamed over SSE.
Python · LangGraph · RAG · SSE
DATA TOOLS / APIFY
Tools for X and YouTube data
Choose a public-data task. Each tool runs on Apify and returns a dataset you can use.
Built by Arjun · Hosted and run on Apify · Independent tools, not affiliated with X or YouTube.
FIELD NOTES
Build notes
Map the Workflow Before Choosing an Agent Framework
A support assistant may need to ask a follow-up question, wait for a reply, and pick up where it left off. Use that workflow to shortlist Agent frameworks, then compare their results, latency, and cost on the same tasks.
The Value of a Multi-Agent System Depends on How You Split the Work
One agent can review a pull request from start to finish, or several agents can check different parts in parallel. The second approach is worth considering when the work can be separated and the results can be handed off clearly.
What Challenges Come With Building a Complex Agent?
A complex Agent has to plan across multiple steps, call tools, recover from failures, and verify its own result. This article breaks down five common engineering challenges through a CI investigation that goes off course.
Why an Agent Uses the Wrong Package Manager in a New Project
New projects default to pnpm, while one older repository still needs npm ci. Both rules are correct, yet the agent picks the wrong tool because Scope disappeared during a Memory write.
When an Agent Forgets, Look Beyond the Model
When an Agent forgets yesterday's instruction, the model may not be at fault. The missing piece is often how the system stores and retrieves memory.
How Agents Plan, Search, and Replan
A useful plan can be executed, checked, and revised. See how decomposition, search, replanning, and formal planners solve different problems.