GenAI PMGemini API
Google’s API for building with Gemini models, including managed agents and developer workflows. In this newsletter it’s highlighted for new agent features like background tasks, remote MCP, function calling, and credential refresh.
Key Highlights
- Gemini API has evolved from model access into a broader platform for agents, retrieval, multimodal generation, and long-running workflows.
- Managed Agents introduced hosted autonomous runtimes with sandboxed code execution, then expanded with background tasks, remote MCP, and function calling.
- File Search and Gemini Embedding 2 make multimodal retrieval over PDFs and images much easier to prototype and ship.
- Webhooks and credential refresh indicate stronger support for production-grade asynchronous and integrated AI workflows.
- For AI PMs, Gemini API is useful both for rapid prototyping in AI Studio and for operationalizing real product workflows.
Gemini API
Overview
Gemini API is Google’s developer platform for building applications and agentic workflows on top of Gemini models. Across the newsletter coverage, it appears not just as a model access layer, but as a broader execution environment for multimodal generation, retrieval, long-running jobs, hosted agent runtimes, and developer tooling integrated with Google AI Studio. For AI Product Managers, that makes it relevant as both a model API and a product surface for shipping end-user features faster.What stands out in recent mentions is the API’s evolution toward production-grade agent development. New capabilities like managed agents, background tasks, remote MCP, function calling, file search, webhooks, and credential refresh suggest Google is positioning Gemini API as a full stack for building reliable AI workflows rather than only issuing prompts to models. For AI PMs, this matters because it reduces the amount of custom infrastructure needed to prototype, validate, and operationalize AI features.
Key Developments
- 2026-04-30: Philipp Schmid shared a getting-started guide for building Deep Research workflows with the Gemini API, covering setup, workflow design, and execution of research queries.
- 2026-05-05: Google introduced webhooks in the Gemini API for long-running tasks, aimed at streamlining workflows such as batch jobs, agents, and media generation.
- 2026-05-06: Logan Kilpatrick launched a multimodal File Search tool in the Gemini API powered by Gemini Embedding 2, adding custom metadata, inline citations, free storage, and on-demand embedding generation.
- 2026-05-07: Philipp Schmid highlighted that Gemini API File Search now supports true multimodal PDF and image retrieval with `gemini-embedding-2`, handling chunking, embedding, indexing, and grounding in a single call.
- 2026-05-12: Philipp Schmid shared Google’s Gemini API interactions quickstart guide to help developers quickly set up and test Gemini models.
- 2026-05-20: Logan Kilpatrick announced major updates across Google AI Studio and the Gemini API, including Gemini 3.5 Flash, managed agents using the antigravity harness, native Android app creation in AI Studio, workspace integrations, and one-click export.
- 2026-05-22: Philipp Schmid demonstrated a GitHub Issue Triage Agent built with a single curl call to the Gemini API, showing how quickly agent workflows could be assembled.
- 2026-06-02: Philipp Schmid launched Managed Agents in the Gemini API, enabling users to start autonomous agents that can reason, write and run code, and manage files in a hosted Linux sandbox with one API call.
- 2026-07-01: Google DeepMind rolled out Gemini Omni Flash via the Gemini API and Google AI Studio for high-quality video generation and editing, alongside Nano Banana 2 Lite as a fast, low-cost Gemini image model.
- 2026-07-08: Logan Kilpatrick released major Managed Agents updates in the Gemini API, adding background tasks, remote MCP, function calling, and network credential refresh, with access available on the free tier.
Relevance to AI PMs
- Prototype agent features without building core infrastructure first. Managed Agents, hosted sandboxes, and function calling let PMs validate use cases like research assistants, triage bots, or workflow automation before committing engineering effort to a custom agent runtime.
- Ship multimodal search and retrieval features faster. File Search with Gemini Embedding 2 reduces the operational burden of document chunking, indexing, metadata handling, and grounding, making it easier to test knowledge assistants over PDFs, images, and internal content.
- Design for real product workflows, not just chat demos. Webhooks, background tasks, remote MCP, and credential refresh are signals that Gemini API supports asynchronous, tool-using, production-oriented experiences such as long-running jobs, enterprise integrations, and agent handoffs.
Related
- Google AI Studio / AI Studio: Closely paired with the Gemini API as Google’s developer environment for testing models, building workflows, and launching app experiences.
- Managed Agents / Managed Agents Quickstart: A major layer on top of the Gemini API that abstracts agent runtime concerns and enables hosted autonomous workflows.
- Background Tasks, Remote MCP, Function Calling: Important agent capabilities recently added to Managed Agents in the Gemini API, expanding orchestration and tool-use options.
- File Search / Gemini Embedding 2: Retrieval components within the Gemini API that support multimodal search, indexing, and grounded responses.
- Gemini 3.5 Flash, Gemini 3 Pro Preview, Gemini 3.1 Flash Lite, Gemini Omni Flash: Model variants surfaced through the Gemini API for different speed, cost, and modality tradeoffs.
- Google DeepMind / Google: The organizations behind the broader Gemini ecosystem and many of the updates referenced in newsletter mentions.
- Philipp Schmid / Logan Kilpatrick: Frequent contributors and evangelists demonstrating new Gemini API workflows, quickstarts, and product launches.
- Deep Research: A workflow category highlighted in relation to Gemini API guides and agentic research use cases.
- GitHub Issue Triage Agent: A concrete example showing how the Gemini API can be used to stand up practical operational agents quickly.
- Google Search / Google Maps: Related Google surfaces that may connect conceptually to broader multimodal and tool-augmented workflows in the Gemini ecosystem.
Newsletter Mentions (21)
“Logan Kilpatrick rolled out major updates to Managed Agents in the Gemini API—adding background task support, remote MCP & function calling, and network credential refresh—and now you can try them on the free tier.”
#2 𝕏 Logan Kilpatrick rolled out major updates to Managed Agents in the Gemini API—adding background task support, remote MCP & function calling, and network credential refresh—and now you can try them on the free tier. Also covered by: @Logan Kilpatrick
“Google DeepMind shipped Nano Banana 2 Lite, its fastest, cheapest Gemini Image model, and rolled out Gemini Omni Flash via the Gemini API and Google AI Studio to enable high-quality video generation and editing.”
Google AI unveiled Gemini Omni, Flash, Nano, and Banana 2 Lite—a suite of multimodal models designed for rapid idea exploration and scalable visual concept creation. Also covered by: @Logan Kilpatrick , @Philipp Schmid , @Google AI #8 𝕏 Google DeepMind shipped Nano Banana 2 Lite, its fastest, cheapest Gemini Image model, and rolled out Gemini Omni Flash via the Gemini API and Google AI Studio to enable high-quality video generation and editing. Also covered by: @Logan Kilpatrick , @Philipp Schmid , @Google AI
“Philipp Schmid launched Managed Agents in the Gemini API, allowing users to spin up autonomous AI agents that reason, write and run code, and manage files inside a hosted Linux sandbox with just one API call.”
#4 𝕏 Philipp Schmid launched Managed Agents in the Gemini API, allowing users to spin up autonomous AI agents that reason, write and run code, and manage files inside a hosted Linux sandbox with just one API call.
“Philipp Schmid built a GitHub Issue Triage Agent using a single curl to the Gemini API.”
#9 𝕏 Philipp Schmid built a GitHub Issue Triage Agent using a single curl to the Gemini API.
“Logan Kilpatrick launched major updates to Google AI Studio and the Gemini API—Gemini 3.5 Flash, managed agents with the antigravity harness, native Android app creation in AI Studio, workspace integrations, and one-click antigravity export.”
#17 𝕏 Logan Kilpatrick launched major updates to Google AI Studio and the Gemini API—Gemini 3.5 Flash, managed agents with the antigravity harness, native Android app creation in AI Studio, workspace integrations, and one-click antigravity export.
“Philipp Schmid shares Google’s Gemini API interactions quickstart guide, helping PM builders quickly set up and test the new Gemini AI model.”
#20 𝕏 Philipp Schmid shares Google’s Gemini API interactions quickstart guide, helping PM builders quickly set up and test the new Gemini AI model. #21 𝕏 Lenny Rachitsky shares eight actionable insights from Eric Ries—spanning financial gravity, CEO retention post-IPO, public-benefit corp structures like AnthropicAI, mission protection, and principled decision-making exemplified by Cloudflare.
“The Gemini API File Search tool now offers true multimodal PDF and image retrieval using `gemini-embedding-2`, handling chunking, embedding, indexing and grounding in one call.”
#4 𝕏 Philipp Schmid : The Gemini API File Search tool now offers true multimodal PDF and image retrieval using `gemini-embedding-2`, handling chunking, embedding, indexing and grounding in one call. #5 𝕏 Google DeepMind partners with EVE Online’s developers to use the game’s complex, player-driven universe as a sandbox for AI agents focused on memory, continual learning, and long-term planning.
“Logan Kilpatrick launched a multi-modal File Search tool in the Gemini API powered by Gemini Embedding 2, now with custom metadata, inline citations, and free storage plus on-demand embedding generation.”
#4 𝕏 Logan Kilpatrick launched a multi-modal File Search tool in the Gemini API powered by Gemini Embedding 2, now with custom metadata, inline citations, and free storage plus on-demand embedding generation.
“Google ships webhooks in Gemini API for long-running tasks #1 𝕏 xAI launched emotion-rich voice cloning on its Grok Voice API, now live for developers to generate AI voices nearly indistinguishable from human speech.”
Google ships webhooks in Gemini API for long-running tasks #1 𝕏 xAI launched emotion-rich voice cloning on its Grok Voice API, now live for developers to generate AI voices nearly indistinguishable from human speech. #2 𝕏 Logan Kilpatrick shipped Webhooks in the Gemini API to streamline developer workflows for long-running tasks like batch jobs, agents, and GenMedia. #3 𝕏 NVIDIA AI launched cuOpt Agent Skills, delivering GPU-accelerated decision optimization for supply-chain planning.
“#10 𝕏 Philipp Schmid published a developer getting-started guide on building and running Deep Research workflows with the Gemini API, covering API setup, workflow construction, and executing deep research queries.”
#10 𝕏 Philipp Schmid published a developer getting-started guide on building and running Deep Research workflows with the Gemini API, covering API setup, workflow construction, and executing deep research queries. #11 𝕏 Cursor launched the Cursor SDK, letting PM Builders spin up agents with the same runtime, harness, and models that power Cursor.
Related
AI developer advocate and AI product communicator associated with Google DeepMind. He is credited here for announcing new Gemini API Managed Agent features.
Google’s AI research lab, mentioned here in connection with interpretability and model reasoning. For PMs, it represents frontier research into understanding and auditing model behavior.
Google AI product leader frequently associated with AI Studio and developer-facing launches. Here he is credited with rolling out GitHub import in AI Studio Build.
Google’s AI assistant/model family, referenced here through Josh Woodward’s community feedback post. The newsletter suggests product improvements are being informed by large-scale user replies.
Technology company named as a challenger in the predicted AI super app market. It is a major platform owner and AI competitor for PMs.
Google’s research organization, mentioned here for launching Open Health Stack and SensorFM. The items suggest work in health infrastructure and wearable-data foundation models.
Google’s app-building environment, here highlighted for globally unique ai.studio subdomains and instant publishing. For PMs, it represents low-friction deployment and branded app distribution.
CEO of Google and Alphabet, mentioned here in connection with Gemini/DiffusionGemma announcements and open-sourcing model weights.
Google model recommended for OCR and VQA workloads. It is highlighted for speed, cost, and accuracy tradeoffs relevant to PM decision-making.
An opinionated build environment for coding with AI that uses a coding agent. The newsletter notes that projects can be exported from it directly to Antigravity.
Google’s search product, mentioned as another interface for detecting SynthID watermarks. It illustrates how AI safety features can be embedded into mainstream consumer search.
A Gemini model variant that was noted as moving out of preview status.
A research capability embedded into Perplexity Computer as a built-in skill. For PMs, it indicates the packaging of advanced research into agent workflows.
An embedding model powering multimodal file search in the Gemini API. Relevant for PMs designing retrieval, citation, and metadata-aware workflows.
Google’s mapping product used as a grounding source in AI Studio. It is mentioned as part of building location-aware, citation-backed apps.
Stay updated on Gemini API
Get curated AI PM insights delivered daily — covering this and 1,000+ other sources.
Subscribe Free