Customizing an LLM for Enterprise Software Engineering

• URL: http://arxiv.org/abs/2605.16517v2
• Date: Tue, 19 May 2026 21:11:07 GMT
• Title: Customizing an LLM for Enterprise Software Engineering
• Abstract summary: We introduce Gemini for Google (GfG), an adaptation of Gemini specialized for Google's internal software engineering ecosystem.<n>In a large-scale blind A/B study across 29,000 developers, Gemini for Google significantly outperformed baselines.<n>We believe this methodology offers a replicable path for other organizations to unlock the full potential of their internal engineering data.
• Score: 35.9413356526649
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Enterprise software development is a continuous evolutionary process, characterized by incremental additions, architectural revisions, production deployments and rigorous maintenance. These activities generate valuable data that modern LLMs could be finetuned on, to unlock additional tool possibilities for enterprise software engineering. While frontier LLMs are already very capable, this form of customization offers a compelling path for enterprise-specific optimization. We introduce Gemini for Google (GfG)}, an adaptation of Gemini specialized for Google's internal software engineering ecosystem. This paper details the model's end-to-end development, from curating a trillion-token proprietary dataset to implementing a mid-training strategy that mitigates catastrophic forgetting. In a large-scale blind A/B study across 29,000 developers, Gemini for Google significantly outperformed baselines: reducing the mean number of iterations per turn by 23\%, and increasing code survival rates by about 17%. Beyond metrics, we provide a comprehensive blueprint for enterprise model adaptation, covering: (1)The extraction of high-value signals from software engineering data, (2)Data preparation strategies, (3)Full-stack model tuning (continued pre-training and post-training), and (4)The deployment of downstream applications. We believe this methodology offers a replicable path for other organizations to unlock the full potential of their internal engineering data.

Related papers

• Exploring Autonomous Agentic Data Engineering for Model Specialization [37.35310287793494]
  Large Language Models (LLMs) have demonstrated strong performance on general tasks, while often struggling to adapt to specialized domains.<n>We formalize textbfAutonomous Agentic Data Engineering, a novel task designed to evaluate LLMs as autonomous data engineers.<n> Experiments show that autonomous LLM data engineers yield substantial gains, as GPT-5.2 constructs a training curriculum that improves a student model by textbf57.29%.
  arXiv  Detail & Related papers  (2026-05-28T17:50:10Z)
• Guiding LLM Post-training Data Engineering with Model Internals from Sparse Autoencoders [49.87696007629591]
  Post-training data engineering largely relies on external signals and ignores rich intrinsic signals lying in model internals.<n>We propose SAERL, a data engineering framework for reinforcement learning (RL)<n>It models three intrinsic data properties: diversity, difficulty, and quality, using model internals extracted with Sparse Autoencoder (SAE), an advanced mechanistic interpretability tool.
  arXiv  Detail & Related papers  (2026-05-26T17:55:59Z)
• Eureka: Intelligent Feature Engineering for Enterprise AI Cloud Resource Demand Prediction [2.101720002743541]
  We present Eureka, an agentic code generation framework for predictive model performance.<n>Eureka has three stages: Expert Agent, Feature Factory, and Self-Evolving Alignment Engine.<n>Eureka consistently outperforms AutoFE and LLM-based baselines on 7 public benchmarks in healthcare, finance, and social domains.
  arXiv  Detail & Related papers  (2026-05-24T23:21:44Z)
• IQuest-Coder-V1 Technical Report [56.045533451719905]
  We introduce the IQuest-Coder-V1 series-(7B/14B/40B/40B-Loop), a new family of code large language models (LLMs)<n>IQuest-Coder-V1 achieves state-of-the-art performance among competitive models across critical dimensions of code intelligence.
  arXiv  Detail & Related papers  (2026-03-17T16:15:31Z)
• SWE-AGI: Benchmarking Specification-Driven Software Construction with MoonBit in the Era of Autonomous Agents [21.8776989802963]
  SWE-AGI is an open-source benchmark for evaluating end-to-end, specification-driven construction of software systems written in MoonBit.<n>Each task involves implementing 1,000-10,000 lines of core logic, corresponding to weeks or months of engineering effort for an experienced human developer.
  arXiv  Detail & Related papers  (2026-02-10T06:31:47Z)
• Towards Reliable ML Feature Engineering via Planning in Constrained-Topology of LLM Agents [1.991571265620589]
  Recent advances in code generation models have unlocked unprecedented opportunities for automating feature engineering.<n>Their adoption in real-world ML teams remains constrained by critical challenges.<n>We address these challenges with a planner-guided, constrained-topology multi-agent framework.
  arXiv  Detail & Related papers  (2026-01-15T19:33:42Z)
• InternBootcamp Technical Report: Boosting LLM Reasoning with Verifiable Task Scaling [71.37579508777843]
  Large language models (LLMs) have revolutionized artificial intelligence by enabling complex reasoning capabilities.<n>To address this gap, we present InternBootcamp, an open-source framework comprising 1000+ domain-diverse task environments.
  arXiv  Detail & Related papers  (2025-08-12T05:00:00Z)
• Thinking Longer, Not Larger: Enhancing Software Engineering Agents via Scaling Test-Time Compute [61.00662702026523]
  We propose a unified Test-Time Compute scaling framework that leverages increased inference-time instead of larger models.<n>Our framework incorporates two complementary strategies: internal TTC and external TTC.<n>We demonstrate our textbf32B model achieves a 46% issue resolution rate, surpassing significantly larger models such as DeepSeek R1 671B and OpenAI o1.
  arXiv  Detail & Related papers  (2025-03-31T07:31:32Z)
• LLM-Generated Microservice Implementations from RESTful API Definitions [3.740584607001637]
  This paper presents a system that uses Large Language Models (LLMs) to automate the API-first development of software.<n>The system generates OpenAPI specification, generating server code from it, and refining the code through a feedback loop that analyzes execution logs and error messages.<n>The system has the potential to benefit software developers, architects, and organizations to speed up software development cycles.
  arXiv  Detail & Related papers  (2025-02-13T20:50:33Z)
• Embedded Software Development with Digital Twins: Specific Requirements for Small and Medium-Sized Enterprises [55.57032418885258]
  Digital twins have the potential for cost-effective software development and maintenance strategies. We interviewed SMEs about their current development processes. First results show that real-time requirements prevent, to date, a Software-in-the-Loop development approach.
  arXiv  Detail & Related papers  (2023-09-17T08:56:36Z)
• MLOps: A Step Forward to Enterprise Machine Learning [0.0]
  This research presents a detailed review of MLOps, its benefits, difficulties, evolutions, and important underlying technologies. The MLOps workflow is explained in detail along with the various tools necessary for both model and data exploration and deployment. This article also puts light on the end-to-end production of ML projects using various maturity levels of automated pipelines.
  arXiv  Detail & Related papers  (2023-05-27T20:44:14Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.