Manoj Mallick | Sigmap

@manojmallick
1 Followers
0 Following
6 Posts

Solution Architect @ ING | Building SigMap — the verification layer for AI coding agents.

🔍 78.9% hit@5 accuracy | 97.9% token reduction | 512 ⭐ on GitHub
Open-source codebase context engine for VS Code, JetBrains & Neovim.

🏗️ 2y production LLM in EU banking | DORA · NIS2 · MiFID II
🌍 Making AI accessible to everyone via LearnHubPlay BV

👉 sigmap.io

Manoj Mallick | Sigmap3d ago

Context Engineering #4

Humans don't read entire repositories.

We navigate using:

• folders
• functions
• references

AI agents should do the same.

SigMap builds a structural map of your codebase for agent navigation.

#ContextEngineering #SigMap

Manoj Mallick | Sigmap3d ago

Context Engineering #3

More context ≠ better answers.

Extra files create noise.

AI agents often fail because relevant code is buried under irrelevant code.

SigMap helps surface the signal before retrieval begins.

#ContextEngineering #AIAgents #SigMap

Manoj Mallick | Sigmap3d ago

Context Engineering #2

You ask:

"How does authentication work?"

The answer may live in 5 files.

Your repo may contain 5,000.

The challenge isn't answering.
It's finding the right context.

SigMap helps map that path.

#ContextEngineering #SigMap

Manoj Mallick | Sigmap4d ago

Context Engineering #1

Prompt engineering asks:
"What should I ask AI?"

Context engineering asks:
"What should AI see?"

Many coding failures happen before generation starts.

That's the problem SigMap is designed to solve.

#ContextEngineering #AIAgents #SigMap

Show threadManoj Mallick | SigmapJun 15

🧵 Why "No vectors. No embeddings"?

Most context tools use vector search — meaning they
embed your entire codebase into a DB and do similarity
lookups at runtime.

Problems:
❌ Slow to index (minutes–hours for large repos)
❌ Embeddings go stale as code changes
❌ Still wrong 20–30% of the time

SigMap uses deterministic file signatures instead.

Result: <50ms lookup. Always fresh. 78.9% accurate.

Try it: sigmap.io
#SigMap #LLM #DevTools #OpenSource

Manoj Mallick | SigmapJun 15

🚀 Introducing SigMap — the verification layer for AI coding agents.

✅ 78.9% hit@5 benchmark accuracy (21 repos tested)
⚡ 97.9% token reduction vs. full codebase ingestion
🔌 VS Code · JetBrains · Neovim plugins + MCP server (9 tools)
📦 ~20K npm downloads | 512 ⭐ GitHub

No vectors. No embeddings. Deterministic. Local. Fast.

👉 sigmap.io
#SigMap #AIAgents #OpenSource #DevTools #buildinpublic