Augment Code
Enterprise AI coding agent for large complex codebases — the Augment Agent and developer extensions ingest entire repositories, build semantic embeddings, and target the agentic-coding lifecycle for software teams managing multi-repository monorepos at scale. April 2024 $227M Series B at ~$977M post-money (Sutter Hill, Index, Lightspeed, Meritech, Innovation Endeavors).
The Business
Augment Code (Augment Inc.) is a Palo Alto-based AI coding agent company founded in 2022 by Scott Dietzen (CEO, former CEO of Pure Storage), Igor Ostrovsky (Co-founder & CTO, ex-Microsoft Azure and Pure Storage) and Guy Gur-Ari (Co-founder & President, ex-Google Brain). The company emerged from stealth in April 2024 with a $227M Series B at a $977M post-money valuation led by Sutter Hill Ventures (Mike Speiser), with Index Ventures, Innovation Endeavors (Eric Schmidt’s fund), Lightspeed Venture Partners and Meritech Capital participating. Evolution Equity Partners added a Series B extension in November 2024, bringing cumulative external capital to approximately $252M. The product line targets enterprise software teams managing large complex codebases: the Augment Agent (200,000-token context, repository-wide semantic indexing, agentic multi-step task execution) plus IDE extensions for VS Code, JetBrains and CLI workflows. The Augment Agent debut in 2025 was anchored on a record-breaking SWE-bench Verified score and a published 70% win rate over GitHub Copilot per VentureBeat coverage.
Customers and Distribution
Augment does not file public financials; the published commercial signals are the 100,000+ developers on the platform cited in company-blog disclosures and the named customer references spanning Webflow, Pure Storage, Observe and Quorum. Distribution sits across three channels: direct enterprise sales to Fortune 500 software-engineering organisations (the principal channel for large-monorepo customers); IDE-extension distribution via the VS Code and JetBrains marketplaces (the bottoms-up developer-acquisition surface); and the Augment Agent CLI for terminal-anchored developer workflows. The customer-acquisition motion has anchored on enterprise codebases too complex for context-limited generalist coding assistants — the 200K-token context and repository-wide semantic indexing are the differentiator the company has consistently emphasised in product messaging and press cycles.
Model Strategy
Augment Code is a Verticals-first play under the IM Framework eight-trajectories taxonomy as applied to coding AI: the strategic bet is that vertical depth on the enterprise-codebase coding primitive beats horizontal generalist coding-AI plays at the largest enterprise software-engineering customers. The model-supplier posture runs on third-party frontier models from OpenAI, Anthropic and others for code generation and agentic reasoning; the differentiated value layer is the codebase-indexing and context-management infrastructure on top. The D4a supplier-diversity sub-rubric was held at 3 in the v1.6ep evidence pass reflecting concentrated supplier exposure relative to the multi-model best practice. The Plus is Plateau: even if frontier-model coding capability plateaus or in-houses into Claude Code / OpenAI Codex first-party products, the enterprise-codebase context infrastructure compounds independently as agentic deployments standardise inside large software organisations.
At A Glance
The Numbers
Annualised revenue
Headcount (FTE)
Funding to date
Leadership Team
Augment Code is led by CEO Scott Dietzen with co-founders Igor Ostrovsky (CTO) and Guy Gur-Ari (President) as the public-facing leadership team. The senior bench draws heavily on Sutter Hill Ventures’ Snowflake / Pure Storage operating playbook and includes Innovation Endeavors (founded by former Google CEO Eric Schmidt) on the cap table. CFO and CRO appointments have not been publicly named at the C-suite level; the company’s careers page is the canonical entry point for senior hires.
IM Framework Scoring
IM’s structured assessment of Augment Code’s competitive position. The summary below is the headline; expand “Show the full analyst-grade analysis” near the bottom for the per-dimension reasoning and evidence. Methodology →
Funding History
| Date | Round | Raised | Post-money | Lead investor(s) |
|---|---|---|---|---|
| Nov 2024 | Series B extension | Undisclosed | — | Evolution Equity Partners |
| Apr 2024 | Series B | $227M | $977M post-money | Sutter Hill Ventures |
| 2023 | Series A | $25M | — | Sutter Hill Ventures |
Cumulative external capital of approximately $252M+ through the April 2024 $227M Series B at $977M post-money led by Sutter Hill Ventures, with Index Ventures, Innovation Endeavors (Eric Schmidt), Lightspeed Venture Partners and Meritech Capital participating; Evolution Equity Partners made a Series B extension investment in November 2024. The April 2024 Series B is the canonical disclosure anchor for the cumulative-capital and valuation figures cited on this page.
Competitive Landscape
| Competitor | Positioning | Distribution edge | Threat profile |
|---|---|---|---|
| GitHub Copilot ((Microsoft)) |
Category-defining AI coding assistant — in-IDE completion, chat, agent mode and the GitHub-native Coding Agent — trained on GitHub’s proprietary corpus and now expanding into autonomous developer surfaces. | Bundled into GitHub Free / Pro / Business / Enterprise plans and pre-installed in VS Code; sold through Microsoft Enterprise Agreements at every Fortune 500 procurement desk. | High — the category incumbent with the broadest enterprise distribution via Microsoft / VS Code / GitHub procurement; the principal structural competitor on enterprise coding-AI procurement. |
| Cursor ((Anysphere)) |
Anysphere’s agent-first AI IDE (VS Code fork) with codebase-aware retrieval, in-editor agents and Composer multi-file editing; positioned as the developer-mindshare leader in 2025-2026. | Direct developer PLG (free / $20 Pro / $40 Business) with viral bottom-up adoption inside engineering teams; reportedly the fastest ARR ramp in the AI-coding cohort. | High — the developer-mindshare leader in 2025–2026 with the fastest ARR ramp in coding AI; head-to-head on the developer-experience surface. |
| Cognition Labs (Devin) | Autonomous AI software engineer positioned as a teammate executing end-to-end tasks (plan, code, run, debug) rather than an IDE assistant — the closest direct competitor on Augment’s agentic positioning. | Direct enterprise pilot sales with seat-based pricing; Slack-integrated workflow plus a hosted Devin environment per developer. | Medium-High — the closest direct competitor on agentic autonomous coding positioning; differentiated on the autonomous-developer framing vs Augment’s enterprise-codebase anchor. |
| Claude Code ((Anthropic)) |
Anthropic’s first-party agentic coding CLI / SDK built on Claude Sonnet / Opus 4.x — the line that leads public coding benchmarks (SWE-bench, Terminal-Bench). Positioned as the model-native way to do agentic engineering. | Anthropic API direct, plus distribution through AWS Bedrock and Google Vertex marketplaces; reaches every Anthropic API customer with zero additional procurement. | High — Anthropic’s first-party agentic coding product line; benefits from the Claude model’s native coding capability lead and Anthropic’s enterprise distribution. |
| OpenAI Codex / ChatGPT coding agents ((OpenAI)) |
OpenAI’s first-party coding agent surface — Codex (cloud agent in ChatGPT) plus GPT-5 / o-series API for coding — positioned as the horizontal default for teams already on the OpenAI stack. | ChatGPT Plus / Pro / Team / Enterprise (Codex in-product), plus OpenAI API and Microsoft Azure OpenAI; rides ChatGPT consumer-to-enterprise expansion through the OpenAI ecosystem. | Medium-High — OpenAI’s first-party coding-agent line bundled into ChatGPT Enterprise and the developer API; structural distribution through the OpenAI ecosystem. |
Potential Risks
Brutal competitive substitution — GitHub Copilot, Cursor, Cognition, Claude Code
Coding AI is the most heavily-contested category in enterprise AI applications: GitHub Copilot’s distribution via Microsoft / VS Code is the structural moat; Cursor’s developer-mindshare lead and ARR ramp are the most active competitive pressure; Cognition’s Devin is the closest direct competitor on agentic autonomous coding; Anthropic’s Claude Code and OpenAI’s Codex / ChatGPT coding agents bring frontier-model-native first-party offerings. The substitution risk is the dominant variable on the score.
Portability of developer tooling caps switching costs
Developer tooling is structurally portable: VS Code extensions, JetBrains plugins and CLI agents can be swapped at low effort, and individual developers churn between Copilot / Cursor / Augment / Claude Code at the project-by-project level. The D1c portability sub-rubric was held at 5 in the v1.6ep evidence pass on this basis; the bull case is that enterprise-codebase context indexing creates a stickier integration than seat-level tooling; the bear case is that the lock-in only takes hold inside the largest monorepo customers.
SWE-bench Verified leaderboard dynamics
Augment Code debuted with a record-breaking SWE-bench Verified score at the Augment Agent launch in 2025, but SWE-bench is a moving target — new submissions from Cognition, Cursor, Anthropic and OpenAI rotate the leaderboard quarterly. The benchmark dynamic is a useful early-stage credibility anchor but not a durable moat; the watched variable is whether Augment can convert benchmark leadership into named-enterprise-customer references at the Fortune 500 tier.
Capital position vs the coding-AI cohort
Cumulative capital of ~$252M through the April 2024 Series B is competitive against most pure-play coding-AI startups but materially smaller than Cursor / Anysphere (multi-billion-dollar raises through 2025–2026) and Cognition Labs (reported large-scale fundraising), and orders of magnitude smaller than Microsoft / Anthropic / OpenAI on first-party coding products. Capital position is a watched variable as the competitive intensity in coding AI accelerates.
Foundation-model supplier dependency
Augment Code runs on third-party frontier models (OpenAI, Anthropic, and others) for code generation and agentic reasoning. The D4a supplier-diversity sub-rubric was held at 3 in the v1.6ep evidence pass reflecting concentrated supplier exposure relative to the multi-model best-practice. The structural argument is that the differentiated enterprise-codebase context indexing layer is the moat regardless of underlying model; the bear case is that frontier-lab first-party coding products eliminate the third-party-agent middle layer over time.
Recent IM Coverage
- Coding AI — sector landing May 2026.
- IM Framework methodology May 2026.
Show recent press coverage of Augment Code
- Apr 2024 — Augment Code raises $227 Million to empower software teams with AI (Augment Code Blog)
- Apr 2024 — Secretive AI coding assistant startup Augment raises $227M to rival GitHub’s Copilot (SiliconANGLE)
- 2025 — Augment Code debuts AI agent with 70% win rate over GitHub Copilot and record-breaking SWE-bench score (VentureBeat)
- Nov 2024 — Evolution Equity Partners Invests In Series B Round of Augment Inc. (PRNewswire)
- Apr 2024 — Code Generation Startup Emerges From Stealth, Raises $227M (AI Business)
Show the source register for the figures on this page
IM operates a primary-source-where-possible discipline. The figures above come from:
- Revenue: Augment Code is private and does not separately disclose revenue or ARR. The April 2024 Series B emergence and subsequent press coverage describe enterprise customer growth but do not publish a specific revenue figure. We decline to publish an ARR number pending a primary disclosure.
- Customer accounts: Public customer references through the Series B cycle and product launches include Webflow, Pure Storage, Observe and Quorum, plus 100K+ developers on the platform per company-blog disclosures. We reference the Series B blog post as the canonical anchor for the developer-count figure.
- Headcount: Augment Code is private and does not disclose headcount in a formal filing. Named-press coverage of the Series B cycle and LinkedIn-visible figures place the company in the low-hundreds range at the Series B cycle. We decline to publish a precise headcount and reference the Augment Code careers page as the canonical entry point.
- Funding to date: Cumulative external capital of approximately $252M through the April 2024 $227M Series B at $977M post-money led by Sutter Hill Ventures with Index Ventures, Innovation Endeavors (Eric Schmidt), Lightspeed Venture Partners and Meritech Capital participating, plus a November 2024 Series B extension from Evolution Equity Partners. Prior $25M Series A also led by Sutter Hill Ventures.
Methodology & Disclaimer
For metric definitions, source-tier hierarchy, and decline-to-publish rules, see the tracker methodology. Confidence dots (• green / • amber / • red) follow the same convention as the AI Tracker.
Spotted a figure you believe is wrong? Send corrections to info@informationmatters.net.
Information Matters Framework scores are the considered opinion of the IM team — human and AI — applied to publicly-available evidence under a disclosed methodology. They are not statements of fact about the companies scored and they are not investment advice.
