[id] ~/srijan.shukla ·
Srijan Shukla.
AI builder/ senior SRE/ 3x founder· shipping practical agents + infra
Building AI agents, reliability tooling, and infra systems that are practical, debuggable, and worth shipping.
Nearly a decade across infrastructure, observability, cloud platforms, and developer tooling. Now focused on AI systems with production discipline.
2026-04-14 Model overthinking is a control problem High thinking budgets can reduce control, especially when stale context and autonomous tool use mix together. → 2026-04-14 Skill Sync Cron The technical problem was simple: one skills repo, two laptops, five agent runtimes. The real problem was that every unstated decision gave the agent room to optimize for control instead of maintainability. → 2026-04-12 Sequented Gated Prompting with a Pi extension How I wanted one cleanup workflow after each feature, tried building a Pi extension for it, and learned that sequential gated prompting is more reliable than bundling everything into one turn. → 2026-04-12 Frontiers in Agentic Design A systems view of modern AI agents: models are only one layer; the real frontier is the harness, context, control loop, memory system, safety boundary, and orchestration around the model. →
2026-04-12 What Happened When I Tried to Use S3 Like Dropbox on macOS I wanted a cheap cold-storage mount that still felt local in Finder. rclone plus FUSE-T got surprisingly close, but the mount flags matter more than most setup guides admit. → 2025-08-27 AWS Cloud Infra Gotchas You can run AWS Batch jobs as pods in your Kubernetes cluster. Pretty cool actually. As expected the AWS tutorial is reliable, you follow it step by step... → 2025-08-27 Dropbox download to your VM Handy script to download all your business from dropbox This script had to undergo a lot of iterations, it failed a lot of times after download for hours on... → 2025-08-27 Geofence validation on Google Maps JS API Note: This post reflects my 2019 exploration of geofence validation. Today, libraries like Turf.js or GEOS handle these cases more efficiently. Suppose you... →
2026-02-20 Chessmaster Grandmaster on your Mac How a nostalgia trip through Windows XP on UTM failed, why QEMU could not provide the Direct3D support Chessmaster needed, and why Whisky ended up being the actual fix. → 2025-08-27 meta, fun, random Reminder: you can just do things a new word pedagogy. which means the method of teaching something. another unrelated word tacit knowledge. knowledge that... → 2025-08-27 you can just do things There is nothing more that needs to be said. → 2025-03-03 speed is king We often underestimate the value of speed in everything we do. People are always amazed when someone does something faster. The combination of speed and... →
[01]2026
alpha
IAM policy rightsizing agent with AI-powered risk signals and instant rollback.
→ [02]2026
chat-harvest
Local-first archive for AI chat traces across agents, designed as a substrate for DSPy/GEPA-style optimization loops.
→ [03]2026
chorus-ai
Frontend-only LLM ensemble interface with bring-your-own OpenRouter/API keys and client-side multi-model workflows.
→ [04]2026
extract-lessons-book
Ebook/document normalizer with optional OCR and LLM-generated advice + flowcharts. It also became a practical OCR evaluation project for me: for most books, plain PDF text extraction worked better than heavy OCR, so GLM OCR ended up as a fallback reserved for noisy pages, scans, or extraction failures.
→ [05]2026
hashlines-mcp
MCP server for hashline-anchored safe file read/edit workflows inspired by this archived write-up: https://web.archive.org/web/20260412162933/https://blog.can.ac/2026/02/12/the-harness-problem/
→ [06]2026
ita-kg
Knowledge-graph representation of the Indian Income Tax Act so agents can traverse section-to-section relationships explicitly.
→