Aim: With the advent of LLMs, sophisticated agentic program repair has become viable at large organizations with large codebases. In this work, we develop an Engineering Agent that fixes the source code based on test failures at scale across diverse software offerings internally. Method: Using Llama as the base, we employ the ReAct harness to develop an agent. We start with a test failure that was triaged by a rule-based test failure bot. We then set up an agentic harness and allow the agent to reason and run a set of 15 actions from reading a file to generating a patch. We provide feedback to the agent through static analysis and test failures so it can refine its solution. We leverage an LLM-as-a-Judge to ensure that the patch conforms to the standards followed by a human review to land fixes. Benchmark Findings: We curated offline benchmarks for our patch generator, the Engineering Agent loop, and the LLM-as-a-Judge. In offline evaluations we found that a specialized 70B model is highly competitive with the much larger but vanilla Llama-405B. In an ablation study, we found that the ReAct harness (neural model) benefited from the symbolic information from static analysis tools and test execution traces. A model that strikes a balance between the solve rate and error rate vs the cost and latency has a benchmark solve rate of 42.3% using an average 11.8 feedback iterations. Production Findings: In a three month period, 80% of the generated fixes were reviewed, of which 31.5% were landed (25.5% of the total number of generated fixes). Feedback from Engineers: We used open coding to extract qualitative themes from engineers' feedback. We saw positive feedback in the form of quick approvals, gratitude, and surprise. We also found mixed feedback when the Engineering Agent's solution was partially correct and it served as a good starting point.

While Multi-modal Large Language Models (MLLMs) have shown impressive capabilities in document understanding tasks, their ability to locate and reason about fine-grained details within complex documents remains understudied. Consider searching a restaurant menu for a specific nutritional detail or identifying a disclaimer in a lengthy newspaper article — tasks that demand careful attention to small but significant details within a broader narrative, akin to Finding Needles in Images (NiM). To address this gap, we introduce NiM-Benchmark, a carefully curated benchmark spanning diverse real-world documents including newspapers, menus, and lecture images, specifically designed to evaluate MLLMs’ capability in these intricate tasks. Building on this, we further propose Spot-IT, a simple yet effective approach that enhances MLLMs capability through intelligent patch selection and Gaussian attention, motivated from how humans zoom and focus when searching documents. Our extensive experiments reveal both the capabilities and limitations of current MLLMs in handling fine-grained document understanding tasks, while demonstrating the effectiveness of our approach. Spot-IT achieves significant improvements over baseline methods, particularly in scenarios requiring precise detail extraction from complex layouts.

Misconfigurations are major causes of software failures. Existing practices rely on developer-written rules or test cases to validate configuration values, which are expensive. Machine learning (ML) for configuration validation is considered a promising direction, but has been facing challenges such as the need of large-scale field data and system-specific models. Recent advances in Large Language Models (LLMs) show promise in addressing some of the long-lasting limitations of ML-based configuration validation. We present the first analysis on the feasibility and effectiveness of using LLMs for configuration validation. We empirically evaluate LLMs as configuration validators by developing a generic LLM-based configuration validation framework, named Ciri. Ciri employs effective prompt engineering with few-shot learning based on both valid configuration and misconfiguration data. Ciri checks outputs from LLMs when producing results, addressing hallucination and nondeterminism of LLMs. We evaluate Ciri's validation effectiveness on eight popular LLMs using configuration data of ten widely deployed open-source systems. Our analysis (1) confirms the potential of using LLMs for configuration validation, (2) explores design space of LLM-based validators like Ciri, and (3) reveals open challenges such as ineffectiveness in detecting certain types of misconfigurations and biases towards popular configuration parameters.

Large cloud providers such as AWS and IBM now provide managed blockchain platforms, showcasing an active interest in blockchains. Unfortunately, blockchains provide poor performance and scalability. This is true even for the Execute-Order-Validate (EOV) style of blockchains which improves over the traditional Order-Execute architecture. We experimentally show that EOV platforms scale poorly using both vertical and horizontal scaling approaches. We find that the throughput is bottlenecked by the Validation and Commit phases, which poorly utilize the resources, limiting performance and scalability.

We consider the problem of inferring the latency sensitivity of user activity in the context of interactive online services. Our method relies on natural experiments, i.e., leveraging the variation in user-experienced latency seen in the normal course. At its core, our technique, dubbed AutoSens, compares the distribution of latency of the user actions actually performed with the underlying distribution of latency independent of whether users choose to perform any action. This then yields a normalized user preference based on latency. We discuss ways of mitigating various confounders and then present our findings in the context of a large online email service, Microsoft Outlook Web Access (OWA).

The rise in popularity of permissioned blockchain platforms in recent time is significant. Hyperledger Fabric is one such permissioned blockchain platform and one of the Hyperledger projects hosted by the Linux Foundation. The Fabric comprises various components such as smart-contracts, endorsers, committers, validators, and orderers. As the performance of blockchain platform is a major concern for enterprise applications, in this work, we perform a comprehensive empirical study to characterize the performance of Hyperledger Fabric and identify potential performance bottlenecks to gain a better understanding of the system. We follow a two-phased approach. In the first phase, our goal is to understand the impact of various configuration parameters such as block size, endorsement policy, channels, resource allocation, state database choice on the transaction throughput & latency to provide various guidelines on configuring these parameters. In addition, we also aim to identify performance bottlenecks and hotspots. We observed that (1) endorsement policy verification, (2) sequential policy validation of transactions in a block, and (3) state validation and commit (with CouchDB) were the three major bottlenecks. In the second phase, we focus on optimizing Hyperledger Fabric v1.0 based on our observations. We introduced and studied various simple optimizations such as aggressive caching for endorsement policy verification in the cryptography component (3x improvement in the performance) and parallelizing endorsement policy verification (7x improvement). Further, we enhanced and measured the effect of an existing bulk read/write optimization for CouchDB during state validation & commit phase (2.5x improvement). By combining all three optimizations1, we improved the overall throughput by 16x (i.e., from 140 tps to 2250 tps).