Alpha Release • Open Source • 100% Local

AI That Grows With You

Stop re-explaining yourself every conversation. Roampal remembers your context, learns what actually works with you, and gets smarter over time—100% local on your machine.

Pre-built: Embedded Python, ready to run • Source: Free, requires toolchain

100% Local
Open Source

Tired of AI That Never Learns From Its Mistakes?

They Don't Learn From Outcomes

Cloud AI remembers what you say—but not what actually worked vs failed with you.

🔄

You Start From Zero Daily

Re-explain your stack, your goals, your context. Every. Single. Day.

💸

Your Data Lives on Their Servers

Memory features behind paywalls. Your conversations stored in the cloud. Forever.

Traditional AI

User: "How do I fix X?"
AI: "Try A"
User: "Didn't work"
AI: "Try A" (again)

Roampal

User: "How do I fix X?"
AI: "Try A"
User: "Didn't work"
AI: "Try B (A failed before)" ✓

Three Ways Roampal Knows You Better Than Any AI

Remembers Who You Are

AI autonomously stores facts about YOU in permanent Memory Bank: your identity, preferences, goals, projects.

You: "I'm building a SaaS in Next.js"
AI: *[stores: identity=founder, stack=Next.js]*
Week later → Always suggests Next.js solutions

Learns Your Knowledge Base

Upload docs, research papers, notes (.txt, .md). Creates searchable knowledge base that persists forever.

Upload: project_architecture.md
Upload: api_conventions.md
→ AI references YOUR docs when answering

Learns What Actually Works

Tracks outcomes for every conversation. Good advice gets promoted. Bad advice gets deleted. Success rates guide future recommendations.

"Try approach A" → Worked! (score +0.2)
"Try approach B" → Failed (score -0.3)
Tracked: A = 90% success, B = 0%
Next time: Suggests A automatically

See It Learn In Real-Time

1

Bad Advice Fails

What's a good restaurant for a date night?

How about that new sushi place downtown?

Went there, wasn't great...

Recording: FAILED Score: 0.5 → 0.2 ↓
2

Good Advice Works

Any other suggestions?

There's that Italian place on Main Street - quiet atmosphere, great reviews.

Perfect! Had an amazing time there.

Recording: WORKED Score: 0.5 → 0.7 ↑
3

Next Time: Instant Recall

Need a restaurant recommendation for tonight

Past Experience
  • Sushi place downtown: Didn't work out
  • Italian place on Main St: Loved it

Based on what you enjoyed last time, I'd suggest the Italian place on Main Street again, or something similar in that area.

What Makes This Different

Pattern Recognition & Learning

Tracks which solutions work with YOUR specific problems. Learns success rates over multiple interactions and surfaces proven patterns automatically.

How It Learns: 1st time: "Try approach A" → works ✓ Tracks: Approach A = 1 success (100%) 2nd time: Similar problem → suggests A → works ✓ → Approach A = 2 successes (100%) You try: "What about approach B?" → fails ✗ Tracks: Approach B = 0% success Next time: "How should I handle this?" AI checks: Approach A 100%, Approach B 0% → Automatically suggests approach A

Watch It Work

See the AI search memory, create notes, and stream responses in real-time. Full transparency into what it's doing.

🔍 Searching memory_bank...
✓ Found 3 relevant memories
💭 Generating response...
[Streams answer token by token]

Cross-Conversation Memory

Remembers your preferences and learned patterns across ALL conversations. No need to repeat yourself.

Current chat: "Database design help"
AI recalls: You prefer PostgreSQL
(learned from past conversations)
→ Suggests Postgres-specific solution

How It Works Under The Hood

Memory Promotion Flow

Working (24h)
Current context
Score ≥0.7 + 2 uses
History (30d)
Recent conversations
Score ≥0.9 + 3 uses
Patterns (∞)
Proven solutions
Permanent storage

Permanent Collections (No Promotion)

Memory Bank
Your preferences & context
LLM-controlled
Books
Reference docs
User-uploaded
Outcome Detection Algorithm

Roampal uses LLM to detect outcomes from natural language:

  • Positive signals: "that worked!", "perfect", "exactly what I needed"
  • Negative signals: "didn't work", "that failed", "not helpful"
  • Implicit signals: Topic changes, abandoning approach, asking similar question
# LLM analyzes conversation and detects outcome
if outcome == "worked":
    score += 0.2  # Positive outcome
elif outcome == "failed":
    score -= 0.3  # Negative outcome
elif outcome == "partial":
    score += 0.05  # Partial success
Knowledge Graph Learning

Roampal builds a Knowledge Graph tracking problem→solution patterns with success rates:

  • Problem Categories: Groups similar problems by concept signatures
  • Solution Tracking: Links solutions to problems with success counts
  • Success Rates: Calculates success_rate = successes / (successes + failures)
  • Pattern Recognition: Surfaces proven patterns before searching memory
# When outcome = "worked", track pattern
problem_concepts = ["deploy", "docker", "production"]
solution_id = "working_abc123"

knowledge_graph["problem_solutions"][signature] = {
  "doc_id": solution_id,
  "success_count": 5,
  "failure_count": 0,
  "success_rate": 1.0  # 5/5 = 100%
}

# Next time you ask about deployment:
# AI finds Docker has 100% success rate
# Kubernetes has 0% success rate
# → Automatically suggests Docker
Privacy Architecture

All data stored locally:

  • Vector DB: ChromaDB (local files)
  • Outcomes: SQLite database
  • Knowledge graph: JSON file
  • LLM: Runs via Ollama (local)

Zero network calls. Zero telemetry.

Cloud Memory vs Local Learning

Cloud AI remembers facts. Roampal learns what works.

Feature Cloud AI Memory Roampal
Memory Type Stores what you say ✓ Learns what works
Outcome Tracking No feedback loop ✓ Scores every result
Bad Advice Handling Stays in memory ✓ Auto-deleted
Pattern Recognition None ✓ Tracks success rates per approach
Failure Avoidance No tracking ✓ Warns about past failures
Privacy Cloud servers ✓ Your machine only
Works Offline No ✓ Yes

Different tools for different needs. Use cloud AI for broad knowledge, Roampal for personal learning that compounds over time.

Who Builds With Roampal?

👨‍💻

Developers

"Remembers my entire stack. Never suggests Python when I use Rust."

  • Learns debugging patterns that work with YOU
  • Recalls past solutions: "This worked 3 weeks ago"
  • Builds knowledge base of your architecture
🎓

Students & Learners

"My personal tutor that remembers what I struggle with."

  • Tracks what concepts you've mastered
  • Adapts explanations to your learning style
  • Builds on previous lessons automatically
✍️

Writers & Creators

"Remembers my story world, characters, and tone."

  • Stores worldbuilding details permanently
  • Tracks character arcs across conversations
  • Learns your writing voice over time
💼

Entrepreneurs & Founders

"My business advisor that knows my entire strategy."

  • Remembers your business model and goals
  • Tracks what marketing approaches worked
  • Builds institutional knowledge as you grow
Private Alpha
Limited Early Access
5-Tier Memory
Learns & Remembers
Zero Telemetry
Fully Private

Frequently Asked Questions

How is this different from cloud AI memory (ChatGPT, Claude, etc.)?

Cloud AI stores what you say. Roampal learns what works.

Three key differences:

  1. Outcome-based learning: Roampal tracks which approaches succeeded vs failed with you, building a pattern library of what actually works for your specific problems. Cloud AI just remembers facts you told it.
  2. Local-first privacy: Your data stays on your machine. No cloud servers, no subscriptions ($9.99 one-time vs $20/month forever).
  3. True ownership: You own your data, your models, your memory. Export everything anytime. No vendor lock-in.
What models does it support?

Multiple model families with tool-calling support:

  • Llama - Meta's models (3B - 70B)
  • Qwen - Alibaba models (3B - 72B)
  • GPT - OpenAI models (20B - 120B)
  • Mixtral - Mistral's mixture-of-experts (8x7B)

All support Roampal's memory tools. Install via Settings → Model Management.

Does it work offline?

Yes! After downloading models, Roampal works completely offline. No internet required.

What are the system requirements?

Minimum: 8GB RAM, 10GB disk space, Python 3.10+, Ollama installed.

Recommended: 16GB RAM, 20GB disk, dedicated GPU for faster inference.

How do I export my data?

All data is stored in plain text files you can access anytime: data/chromadb/, data/knowledge_graph.json, data/sessions/. Copy the entire folder to backup.

Why charge if it's open source?

The code is free on GitHub. The $9.99 gets you:

  • Pre-packaged executable - No Python setup required
  • Bundled dependencies - Everything included
  • Tested build - Guaranteed to work
  • Lifetime updates - All future releases included

Technical users can build from source for free. The $9.99 is for convenience.

Is Roampal's advice always reliable?

No. Roampal uses AI models that can generate incorrect information, hallucinate facts, or reflect training biases.

Always verify critical information from authoritative sources. Do not rely on AI-generated content for:

  • Medical, legal, or financial advice
  • Safety-critical systems or decisions
  • Production code without thorough review

Why Roampal Exists

Roampal is an experiment in building sustainable technology companies without artificial scarcity, surveillance capitalism, or bait-and-switch licensing.

Ready to Try AI That Actually Learns?

Get instant access to AI that learns and grows with you. Your data stays on your machine.

Pre-built: No dependencies needed • Or build from source (free, requires Node/Rust/Python)