name : Sravya Vedantham
role : AI Engineer · Agentic AI · Video Generation Research
location : United States
focus : autonomous agents, multi-agent orchestration, AI-powered video pipelines
writing : 33 articles on Dev.to — AI, Python, Rust, Privacy, DevTools
status : open to new ideas & collaborations
belief : "Build tools that think for themselves — then open-source them."secret-time-machine — scans your entire git history for secrets that were committed and "deleted". The key you removed in 2023 is still in commit
a4f8c2d. This tool finds it.
pip install secret-time-machine
Tools that change how AI agents think, act, and deploy.
| tool | what it does |
|---|---|
| assemble | multi-agent orchestrator — assembles AI teams, runs them in parallel with wave execution |
| mem-mesh | universal AI memory layer — drop-in proxy that gives Claude persistent memory across conversations |
| guardian | autonomous code guardian — Claude agent that scans your repo and opens PRs with fixes |
| llm-guard | AI agent bodyguard — catches leaked keys, runaway loops, and budget burns before they happen |
| llm-lens | flight recorder for AI agents — replay every decision an agent made |
| llm-bench | LLM benchmarking tool — side-by-side model comparison with real metrics |
| nlops | natural language DevOps — plain English to production-ready Terraform |
| skillforge | AI skill compiler — plain English workflows to production-ready agents |
| clipforge | AI video pipeline — raw gameplay to TikTok/YouTube/trailer clips automatically |
| tool | what it does |
|---|---|
| shadowscan | scans what AI can see about you — env vars, SSH keys, clipboard, git history |
| datawipe | automates GDPR/CCPA right-to-erasure requests for 55+ companies |
| consentmap | privacy policy scanner — plain-English risk summary, no LLM required |
| ai-opt-out | generates opt-out requests to 15 AI companies training on your data |
| trackmap | maps which companies track your browsing — scans history, finds 72 trackers |
| phantom | finds ghost accounts on breached services you forgot about |
| leakwatch | continuous breach monitoring with action tracking and severity scoring |
| privacykit | all three above as a single Rust binary — trackmap + phantom + leakwatch |
| inboxscan | scans Gmail for forgotten subscriptions burning your money |
| tool | what it reveals |
|---|---|
| git-personality | your D&D alignment from commit history |
| hallucination-grep | functions your LLM invented that don't exist |
| token-diet | wasted tokens burning your LLM budget |
| llm-model-diff | exactly where GPT-4 and Claude disagree |
| dev-dna | your developer archetype from GitHub history |
| commit-prophet | which files will have the next bug |
| vibe-check | how much of your codebase an AI wrote |
| prompt-rs | type-safe LLM prompts in Rust — catch prompt bugs at compile time |
| ctx-kit | 5MB Rust binary between your code and every LLM API — cut your bill by 40% |
33 articles · dev.to/lakshmisravyavedantham
latest
- Google Can See 78% of Your Browsing History — I Built a Tool to Prove It
- You Have Accounts on 34 Services You Forgot About. 8 Were Breached.
- I Built a Breach Monitoring System That Runs in Your Terminal
- I scanned my Gmail and found $91/mo in subscriptions I completely forgot about
- I built a Claude Code skill that assembles AI teams and runs them in parallel
