TaceoLabs/co-snarks
Nov 2024 – Jul 2025
5 Months active
Languages Used
CircomJavaScriptNoirRustShellTOMLYAMLC++
Technical Skills
ACVMAPI DesignBackend DevelopmentBinary DevelopmentBuild ConfigurationCI/CD
Franco Nieddu
PROFILE
Franco Nieddu
Over five months, Nieddu contributed to TaceoLabs/co-snarks by engineering scalable, high-throughput zero-knowledge proof infrastructure. He implemented parallelized Groth16 reductions in libsnark using Rust and rayon, enabling multi-core polynomial commitment evaluation. Nieddu also refactored witness extension logic to reuse compiled circuit data, reducing re-compilation overhead and improving MPC VM throughput. His work on batched SumCheck protocols and parallel msm_public_points computation enhanced performance and scalability for large-scale proofs. Throughout, he applied deep expertise in cryptography, Rust, and distributed systems, delivering robust code with careful attention to maintainability, correctness, and efficiency across complex circuit execution and virtual machine workflows.
Overall Statistics
Feature vs Bugs
86%Features
Repository Contributions
71Total
Bugs
6
Commits
71
Features
36
Lines of code
21,140
Activity Months5
Your Network
10 peopleSame Organization
@taceo.io1
Work History
July 2025
1 Commits • 1 Features
Jul 1, 2025Month: 2025-07 — Highlights include delivering a parallelized Groth16 reduction in libsnark using the rayon crate for data parallelism, which enables multi-core evaluation and multiplication of polynomial commitments and improves the performance of the Groth16 proof system. Commit referenced: 3b2b4e0e46410719cf91294b2629406396e8aa11. Major bugs fixed: none reported this month. Overall impact: enhances throughput and scalability of zk-SNARK workflows, enabling faster proof generation on multi-core CPUs and better resource utilization. Technologies/skills demonstrated: Rust, rayon-based parallelism, multithreading patterns in cryptographic computations, libsnark integration, and careful change-management via a focused commit."
1 Commits • 1 Features
Jul 1, 2025Month: 2025-07 — Highlights include delivering a parallelized Groth16 reduction in libsnark using the rayon crate for data parallelism, which enables multi-core evaluation and multiplication of polynomial commitments and improves the performance of the Groth16 proof system. Commit referenced: 3b2b4e0e46410719cf91294b2629406396e8aa11. Major bugs fixed: none reported this month. Overall impact: enhances throughput and scalability of zk-SNARK workflows, enabling faster proof generation on multi-core CPUs and better resource utilization. Technologies/skills demonstrated: Rust, rayon-based parallelism, multithreading patterns in cryptographic computations, libsnark integration, and careful change-management via a focused commit."
July 2025
June 2025
1 Commits • 1 Features
Jun 1, 2025June 2025: Delivered a feature to reuse compiled circuit data across witness extensions in TaceoLabs/co-snarks, enabling the MPC VM to reference CoCircomCompilerParsed instead of consuming it. The change clones several fields to preserve data integrity and support multiple uses of the parsed circuit data, reducing re-compilation overhead and improving throughput for MPC workloads. This work lays groundwork for future caching and shared data usage across VM components, delivering measurable business value in efficiency and reliability.
June 2025
1 Commits • 1 Features
Jun 1, 2025June 2025: Delivered a feature to reuse compiled circuit data across witness extensions in TaceoLabs/co-snarks, enabling the MPC VM to reference CoCircomCompilerParsed instead of consuming it. The change clones several fields to preserve data integrity and support multiple uses of the parsed circuit data, reducing re-compilation overhead and improving throughput for MPC workloads. This work lays groundwork for future caching and shared data usage across VM components, delivering measurable business value in efficiency and reliability.
March 2025
21 Commits • 8 Features
Mar 1, 2025March 2025 (2025-03) monthly summary for TaceoLabs/co-snarks: This period delivered scalable batched computation capabilities, rigorous correctness work, and meaningful maintenance improvements that collectively increase throughput, reliability, and maintainability. Key refactors simplified linear operations in co-noir, removed self references, improved naming, and introduced a clearer brillig success struct. Commit highlights include cf1a62b4910e4bb369d61078332514fdc308eada; c32fd9043e0397736a809b3059ae862522a7abe1; e0af901e2999cc7e38215f36fe2a647b18d94e0e. Major batched features and extensions were delivered: batched sum check with edge extension in 2^20 batches; ultra arith relation now uses MT and half shares; ACVM stores values and returns circuit outputs; batched WTN extension for plain witness plus ChaCha tests; rep3 batched WTN extension for ChaCha; batched ChaCha operations; and run-and-return net for batched WTN extension. Commit lines include f81597601bbb5ee9b501cecd61b479425f05ebc0; d7a6b9a319124204a8fa9d94e4197a6dbf08ede4; 3df88fb244b191e03bbd6e6aaede86eaaf7f3d6b; 36e69cc7621b8689e3c829c8f1489344a1298899; 310a5dc09fc93ab6070571bbe509097817bf2979; a4cb900128dc231660623f16a4fdc02cf181dc10; f299c85ef8f368baa9f90a904ec65942f110c41b. Reliability and correctness improvements were achieved through fixes for large integer divisions in the plain driver and support for multiple return values in Circom WTN S extension from the compute bucket. Commits: a87cac70c7ac9e465d5c8e3a5b6b5fa25ab35beb; 06bc96b7048a95756ba535c23d132fc9eae369a5. Performance benchmarking extended with batched benchmarks: batched_add benches and updates to local_mul_vec bench. Commit: 8ca4cd8949277a09d466c5a2bee98cf10c4a7cbc. Quality, tests, and documentation saw targeted improvements: rep3 ChaCha batched tests; docs for batched WTN extension; code cleanup removing redundant clones, trace logs, and unused imports; minor test crate updates and docs fixes. Commits: 5fb28bd785687b94da1ecb54ee5e54aeb5adaa72; a0670298b1b05a6dc63c1f6d61bec2e8972b1933; c53e5d80439ef1a9926a318622ebd1adcdb502f3; 93ea1ebb8d35736870e494ea20a0104426dbb4ee; 8b81675261fc255ed962f0a35e86a3b3cf6045fb; a37e4d2950691ed3fe877c556988080ddbbdbdbc.
21 Commits • 8 Features
Mar 1, 2025March 2025 (2025-03) monthly summary for TaceoLabs/co-snarks: This period delivered scalable batched computation capabilities, rigorous correctness work, and meaningful maintenance improvements that collectively increase throughput, reliability, and maintainability. Key refactors simplified linear operations in co-noir, removed self references, improved naming, and introduced a clearer brillig success struct. Commit highlights include cf1a62b4910e4bb369d61078332514fdc308eada; c32fd9043e0397736a809b3059ae862522a7abe1; e0af901e2999cc7e38215f36fe2a647b18d94e0e. Major batched features and extensions were delivered: batched sum check with edge extension in 2^20 batches; ultra arith relation now uses MT and half shares; ACVM stores values and returns circuit outputs; batched WTN extension for plain witness plus ChaCha tests; rep3 batched WTN extension for ChaCha; batched ChaCha operations; and run-and-return net for batched WTN extension. Commit lines include f81597601bbb5ee9b501cecd61b479425f05ebc0; d7a6b9a319124204a8fa9d94e4197a6dbf08ede4; 3df88fb244b191e03bbd6e6aaede86eaaf7f3d6b; 36e69cc7621b8689e3c829c8f1489344a1298899; 310a5dc09fc93ab6070571bbe509097817bf2979; a4cb900128dc231660623f16a4fdc02cf181dc10; f299c85ef8f368baa9f90a904ec65942f110c41b. Reliability and correctness improvements were achieved through fixes for large integer divisions in the plain driver and support for multiple return values in Circom WTN S extension from the compute bucket. Commits: a87cac70c7ac9e465d5c8e3a5b6b5fa25ab35beb; 06bc96b7048a95756ba535c23d132fc9eae369a5. Performance benchmarking extended with batched benchmarks: batched_add benches and updates to local_mul_vec bench. Commit: 8ca4cd8949277a09d466c5a2bee98cf10c4a7cbc. Quality, tests, and documentation saw targeted improvements: rep3 ChaCha batched tests; docs for batched WTN extension; code cleanup removing redundant clones, trace logs, and unused imports; minor test crate updates and docs fixes. Commits: 5fb28bd785687b94da1ecb54ee5e54aeb5adaa72; a0670298b1b05a6dc63c1f6d61bec2e8972b1933; c53e5d80439ef1a9926a318622ebd1adcdb502f3; 93ea1ebb8d35736870e494ea20a0104426dbb4ee; 8b81675261fc255ed962f0a35e86a3b3cf6045fb; a37e4d2950691ed3fe877c556988080ddbbdbdbc.
March 2025
February 2025
8 Commits • 3 Features
Feb 1, 2025February 2025: Delivered performance-focused enhancements to co-snarks, focusing on batched SumCheck processing, parallel msm_public_points computation, and build-time cleanup. Implemented batched relations in SumCheck for MPC friendliness, eliminated old unbatched paths, and introduced skip logic to prune non-critical rows. Refactored msm_public_points to use rayon::join for parallel execution of two msm_bigint operations, and adjusted minimum chunk size. Removed timing logs and the tokio dependency to streamline the build and reduce runtime overhead. These changes improved throughput for MPC workflows, enhanced scalability of relation computations, and reduced maintenance overhead.
February 2025
8 Commits • 3 Features
Feb 1, 2025February 2025: Delivered performance-focused enhancements to co-snarks, focusing on batched SumCheck processing, parallel msm_public_points computation, and build-time cleanup. Implemented batched relations in SumCheck for MPC friendliness, eliminated old unbatched paths, and introduced skip logic to prune non-critical rows. Refactored msm_public_points to use rayon::join for parallel execution of two msm_bigint operations, and adjusted minimum chunk size. Removed timing logs and the tokio dependency to streamline the build and reduce runtime overhead. These changes improved throughput for MPC workflows, enhanced scalability of relation computations, and reduced maintenance overhead.
November 2024
40 Commits • 23 Features
Nov 1, 2024November 2024 monthly highlights for TaceoLabs/co-snarks focused on stability, performance, and extended functionality. Delivered deterministic build reproducibility through targeted Circom dependency pinning across the batch, laid groundwork for accelerator integration, expanded Brillig/VM capabilities, and reinforced testing and documentation. Key improvements also include safer zkey and prover workflows with enhanced error handling and curve-check configurability, plus robust CI/testing infrastructure.
40 Commits • 23 Features
Nov 1, 2024November 2024 monthly highlights for TaceoLabs/co-snarks focused on stability, performance, and extended functionality. Delivered deterministic build reproducibility through targeted Circom dependency pinning across the batch, laid groundwork for accelerator integration, expanded Brillig/VM capabilities, and reinforced testing and documentation. Key improvements also include safer zkey and prover workflows with enhanced error handling and curve-check configurability, plus robust CI/testing infrastructure.
November 2024
Activity
Loading activity data...
Quality Metrics
Correctness87.6%
Maintainability86.6%
Architecture85.4%
Performance82.6%
AI Usage20.8%
Skills & Technologies
Programming Languages
C++CircomGoJavaScriptNoirPythonRustShellTOMLYAML
Technical Skills
ACVMAPI DesignAlgorithm optimizationBackend DevelopmentBenchmarkingBig Integer ArithmeticBinary DevelopmentBuild ConfigurationC++CI/CDCargoCircomCircuit DesignCircuit ExecutionClippy
Repositories Contributed To
Overview of all repositories you've contributed to across your timeline