List of topics, talks, workshops and ideas.

Unlocking Incentivized Retrievals with FilBeam

Virtual talk by Miroslav - due by Oct 16-17.

What are the goals of your talk?

• Reignite discussions about the importance of retrievals in Filecoin.

• Highlight the challenges and why retrievals are still hard to solve.

• Introduce Filecoin Beam as a fresh approach bringing incentives to the retrieval layer.

• Explore how developers and creators can build on FilBeam today and what new use cases it will unlock in the near future.

What is the description of your talk?

Retrievability has always been Filecoin’s biggest open problem. Persistence alone doesn’t make a storage network usable—without retrievals, there’s no complete storage solution. In this talk, I’ll trace the history of retrieval attempts in Filecoin, from early whitepaper ideas to experiments like payment channels and Saturn, and explain why retrievals remain hard to solve.

Then I’ll introduce FilBeam — a new approach that brings incentives directly to the retrieval layer. By rewarding Storage Providers for egress, FilBeam turns retrievals into a first-class, sustainable part of the Filecoin economy. More than just enabling reliable access, it unlocks new opportunities: monetising access with x402, enabling agents and scrapers to pay for data access programmatically, and creating revenue streams for data set producers.

By the end of this talk, you’ll see how incentives can finally close the retrievability gap and why FilBeam may be the missing piece that makes Filecoin a complete storage offering.

Short CFP blurb

Filecoin’s future depends on retrievals. This talk dives into why achieving reliable retrievability is so challenging, the lessons we’ve learned, and how FilBeam introduces incentives to make retrievals sustainable. What’s more, FilBeam turns retrievals into opportunities by enabling pay-per-access, agentic data flows, and new income for dataset creators.

Private notes

• History - Filecoin Whitepaper, impossibility of fair exchange, payment channels, clients not being used to pay for retrievals directly, Saturn, etc.

• Future directions - pay for access is becoming accepted - x402, agents, scrapers.

• FilBeam future:
  • FilBeam x402 making it super easy to monetize access to data stored on Filecoin.
  • MCP server allowing LLM agents to work with datasets stored on Filecoin, while compensating dataset producers for their work.

Speaker Bio - Miroslav

Miroslav Bajtoš is the CTO and Co-Founder of Space Meridian. He focuses on solving the retrievability challenges of the Filecoin network and explores how retrievals can become a sustainable, first-class part of decentralised storage. Before founding Space Meridian, Miroslav worked at Protocol Labs in the Retrieval Markets Lab. He led Filecoin Spark, which made retrieval requests 10× more likely to succeed. He also worked on Filecoin Station, which enabled non-technical users to earn FIL by sharing their idle computing resources.

Verifiable discovery of content stored on Filecoin

In-person workshop session led by Miroslav in Buenos Aires (mid-November)

What are the goals of your talk?

• Reach a rough consensus on the approach for verifiable discovery of Filecoin content (Filecoin Pieces and individual IPFS blocks).

• Explore how to provide verifiable mappings from Pieces → IPFS blocks → IPNI advertisements.

• Define a high-level plan for implementing the chosen approach, with concrete next steps and DRIs assigned.

What is the description of your talk?

This workshop will bring together core engineers from FilOz, IPNI, FilBeam, and IPFS to address the challenge of verifiable discovery of content stored on Filecoin. Today, components such as FilBeam cannot reliably discover all IPFS block CIDs contained within a given Filecoin Piece. At the same time, while Curio scans Piece bytes to discover IPFS blocks and advertises these blocks to IPNI, there is currently no way for clients or third parties to verify that these advertisements are complete and correct.

The session is designed for a small group of senior engineers and decision-makers. We will explore approaches for:

• Enabling third parties like FilBeam to discover all IPFS block CIDs contained within Filecoin Pieces.

• Verifying that the complete and correct list of IPFS blocks is advertised to IPNI.

• Allowing IPFS and Filecoin clients to discover retrieval options at both Filecoin Piece and IPFS Block levels.

Designing x402 on Filecoin Pay and Pay-to-Access Retrievals for Filecoin

In-person workshop session led by Miroslav in Buenos Aires (mid-November)

What is the description of your session?

This 60-minute workshop will bring together engineers from FilOz, FilBeam, and Curio to co-design an MVP for pay-to-access retrievals on Filecoin, powered by x402-compatible micropayments on Filecoin Pay. The session will focus on two layers of innovation:

1. x402 on Filecoin Pay — establishing a general-purpose payment rail for off-chain, signed microtransactions that can be settled trustlessly on-chain.

2. Pay-to-access for content retrievals — integrating this mechanism into FilBeam and Curio to enable retrieval clients to pay Storage Providers per request, using signed payment tickets and ZK rollups for settlement.

We will collaborate to validate the proposed design’s technical feasibility, outline a cross-team MVP implementation plan, and identify DRIs for each subsystem. The pre-read document will cover the core architecture and open questions such as ticket nonces, rollup design, and cache-miss payments. The session will be a focused, single-group discussion aimed at producing a written summary of the MVP scope, key decisions, and immediate next steps.

What are the goals of your session?

1. Validate the technical feasibility of implementing x402 micropayments on Filecoin Pay and integrating them into FilBeam and Curio.

2. Define a high-level MVP scope and implementation plan across participating teams.

3. Identify DRIs and immediate next steps for delivering the MVP.

Pre-reading:

🎴Pay-to-access via x402 on Filecoin Pay