Skip to content

bounties.md

Latest commit

 

History

History
135 lines (97 loc) · 3.47 KB

bounties.md

File metadata and controls

135 lines (97 loc) · 3.47 KB

MIND Bounty Board

This page tracks high-impact issues eligible for bounties. Bounties are awarded for merged PRs that fully address the listed requirements.

How Bounties Work

  1. Find an issue marked with 💰 on this page
  2. Comment on the issue to claim it (first come, first served)
  3. Submit a PR that meets the acceptance criteria
  4. Get reviewed by core team
  5. Receive bounty after PR is merged

Current Bounties

💰 $500 - Implement MLIR Tensor Dialect Integration

Issue: [#TBD]
Difficulty: Advanced
Skills: Rust, MLIR, compilers

Description: Build the bridge between MIND's type system and MLIR's tensor dialect.

Requirements:

  • Map MIND tensor types to MLIR tensor types
  • Implement shape inference in MLIR
  • Handle device placement annotations
  • Add comprehensive tests
  • Document the implementation

Timeline: 4-6 weeks

💰 $300 - Build Shape Checker with Error Messages

Issue: [#TBD]
Difficulty: Medium
Skills: Rust, type systems, compiler diagnostics

Description: Implement compile-time shape checking with helpful error messages.

Requirements:

  • Implement shape unification algorithm
  • Add symbolic shape support (e.g., batch size B)
  • Generate clear error messages with suggestions
  • Add 50+ test cases covering edge cases
  • Write developer documentation

Timeline: 3-4 weeks

💰 $200 - Create VSCode Extension

Issue: [#TBD]
Difficulty: Medium
Skills: TypeScript, VSCode API, language servers

Description: Build syntax highlighting and basic LSP support for MIND in VSCode.

Requirements:

  • Syntax highlighting for all MIND constructs
  • Basic LSP server with:
    • Go to definition
    • Hover information
    • Error diagnostics
  • Code snippets for common patterns
  • Published to VSCode marketplace

Timeline: 2-3 weeks

💰 $150 - Implement Autodiff for Loops

Issue: [#TBD]
Difficulty: Advanced
Skills: Rust, automatic differentiation, compilers

Description: Extend autodiff to work through for and while loops.

Requirements:

  • Implement loop unrolling for static bounds
  • Add tape-based recording for dynamic bounds
  • Handle break/continue statements
  • Add benchmarks comparing with PyTorch
  • Write technical documentation

Timeline: 3-4 weeks

💰 $100 - Write Comprehensive Benchmarks Suite

Issue: [#TBD]
Difficulty: Low-Medium
Skills: Rust, benchmarking, MIND

Description: Create a comprehensive benchmark suite comparing MIND with PyTorch/JAX.

Requirements:

  • Implement 10+ common ML operations
  • Equivalent implementations in PyTorch and JAX
  • Run on CPU and GPU
  • Generate comparison graphs
  • Document methodology

Timeline: 2 weeks

Bounty Rules

Eligibility

  • Open to everyone
  • Can claim only one bounty at a time
  • Must not be a core team member (for paid bounties)

Process

  1. Bounties are paid via GitHub Sponsors, Open Collective, or crypto
  2. Payment within 2 weeks of PR merge
  3. Disputes resolved by core team vote

Quality Standards

  • Code must pass CI (tests, clippy, formatting)
  • Must include tests and documentation
  • Must follow contribution guidelines
  • Breaking changes require RFC approval

Proposing New Bounties

Open a discussion with problem, acceptance criteria, difficulty, timeline, and suggested amount.

Sponsoring Bounties

Last Updated: 2025-11-06
Total Active Bounties: $1,250