On-chain activity
Carbon
Carbon is an indexing framework for Solana that processes account updates and transaction notifications through a modular pipeline architecture, enabling developers to maintain program account indexes and build real-time protocol monitors with configurable data sources.
SevenLabs news, features & analysis
Matched from published articles, podcasts, and talks using the project name, token name, or token symbol.
SevenLabs
SevenLabs is a Solana engineering firm that designs and builds production-grade blockchain infrastructure — from on-chain programs and custom indexers to high-performance Rust backends and transaction systems — for teams working at the protocol and application layers of the Solana ecosystem.
What SevenLabs Does
The firm describes its core focus as solving "hard Solana engineering problems," offering four service lines:
Smart contracts — compute-optimized, security-first programs written with production patterns and compliance considerations, targeting protocol-level and application-layer deployments on Solana.
Custom indexing — real-time and historical data infrastructure designed for high availability and horizontal scale. SevenLabs designs bespoke indexing pipelines that capture account changes, transactions, and program events from the Solana ledger and route them to downstream storage or analytics systems.
Rust backends — high-performance APIs engineered for the throughput demands typical of DeFi protocols, NFT marketplaces, and infrastructure providers operating at scale.
Infrastructure — systems for fast transaction submission and low-latency access to on-chain data, including Geyser plugin integrations and RPC infrastructure tuning.
The firm lists fifteen clients, including the Solana Foundation, Ergonia, Bags, Eco, Caldera, Glyde, Hype, Lightcone, and Clicker. SevenLabs is headquartered in Dubai, UAE.
Carbon: Open-Source Indexing Framework
The most visible output of SevenLabs' engineering work is Carbon, an open-source Rust framework for building Solana indexers and data pipelines (MIT license; github.com/sevenlabs-hq/carbon).
Carbon is organized around a three-stage modular pipeline:
Datasources stream Solana updates — transactions, account changes, account deletions, and block metadata — into the runtime. Supported ingestion methods include Geyser plugin streams for low-latency production indexers, standard Solana RPC endpoints for simpler setups, Helius Laserstream, a Jetstreamer datasource, validator snapshot sources for loading current account state, and historical snapshot sources for backfills and range replays. The adapter datasource pattern supports custom message streams from any provider.
Decoders transform raw account and transaction data into typed Rust structures. Decoder crates cover widely used Solana programs across core SPL programs, DeFi protocols, NFT infrastructure, and general Solana tooling. They support instruction decoding, nested CPI call decoding, and emitted event extraction. IDL-based decoder generation from Anchor or Codama specifications is available through a CLI tool, allowing developers to generate typed decoders for any IDL-compatible on-chain program.
Processors receive decoded data and route it to target systems. Built-in helpers expose data through Postgres for typed persistence, GraphQL via Juniper and Axum, Prometheus for metrics, and standard logging. All components are interchangeable: developers can replace RPC streams with Geyser streams, swap generated decoders for custom implementations, or direct output to any storage backend, analytics system, bot, or application-specific sink.
Carbon's design deliberately avoids coupling between stages. A team can combine a public RPC datasource with a custom decoder and a Postgres processor without modifying any other part of the pipeline, making it practical for projects ranging from quick prototypes to high-throughput production indexers.
Adoption and Ecosystem Impact
As of mid-2026, Carbon has accumulated over 611 GitHub stars and 183 forks, with the project reporting more than 100,000 downloads. The framework is referenced in official Solana developer documentation and resources. Bware Labs maintains an independent fork (github.com/bwarelabs/solana-carbon), indicating adoption beyond SevenLabs' direct client base.
Carbon reached v1.0.0 with contributions from twelve individuals, including six new contributors to the repository. Across eighteen releases, the project has added datasource support for providers including Helius, Jetstreamer, and Yellowstone gRPC, decoder crates for protocols including Meteora, Raydium, Bonkswap, and PancakeSwap, and infrastructure for typed Postgres persistence, GraphQL APIs, and block-only subscriptions.
SevenLabs presented Carbon at Solana Accelerate NYC 2025, where Kellian Vitre delivered a talk titled "Scale or Die at Accelerate 2025: Indexing Solana Programs with Carbon," demonstrating how the framework handles indexing at production scale.
Ecosystem Fit
SevenLabs occupies the developer infrastructure layer of the Solana ecosystem, providing both paid engineering services and open-source tooling. For teams that need to build custom indexers without writing pipeline boilerplate, Carbon provides a production-tested Rust framework with pre-built decoders for the most commonly used Solana programs. For teams that need end-to-end engineering capacity — from on-chain programs through backend APIs — SevenLabs offers direct engagement.
The firm also describes a teaching component to its work, positioning itself as a resource for teams learning to build on Solana alongside its commercial engineering services.
Contents
- What SevenLabs Does
- Carbon: Open-Source Indexing Framework
- Adoption and Ecosystem Impact
- Ecosystem Fit
Solana Token Markets
