{"schema_version":"onlylabs.public_analysis.v1","url":"https://onlylabs.fyi/analysis/together-ai","json_url":"https://onlylabs.fyi/analysis/together-ai/analysis.json","evidence_json_url":"https://onlylabs.fyi/analysis/together-ai/evidence.json","generated_at":"2026-06-27T22:31:49.791Z","analysis":{"org_slug":"together-ai","url":"https://onlylabs.fyi/analysis/together-ai","json_url":"https://onlylabs.fyi/analysis/together-ai/analysis.json","evidence_json_url":"https://onlylabs.fyi/analysis/together-ai/evidence.json","dossier_url":"https://onlylabs.fyi/labs/together-ai","org":{"slug":"together-ai","name":"Together AI","category":"neocloud","category_label":"Neocloud","homepage_url":"https://www.together.ai"},"title":"Together AI analysis","summary":"Together AI is consolidating its positioning as the AI-native cloud — an inference-first infrastructure platform that competes on raw speed and cost per token. The evidence pack shows the company simultaneously building out in three directions: (1) deepening the infrastructure surface from GPU clusters into managed storage, networking, and observability, (2) layering enterprise trust and access-control primitives…","markdown":"## Thesis\n\nTogether AI is consolidating its positioning as the AI-native cloud — an inference-first infrastructure platform that competes on raw speed and cost per token. The evidence pack shows the company simultaneously building out in three directions: (1) deepening the infrastructure surface from GPU clusters into managed storage, networking, and observability [P3](https://job-boards.greenhouse.io/togetherai/jobs/5172169007)[E23](https://job-boards.greenhouse.io/togetherai/jobs/5155722007), (2) layering enterprise trust and access-control primitives (ISO 27001 certification, identity and RBAC systems) [P24](https://www.together.ai/blog/iso-27001-2022-certification)[P2](https://job-boards.greenhouse.io/togetherai/jobs/5169470007)[P13](https://job-boards.greenhouse.io/togetherai/jobs/5162910007), and (3) investing heavily in research that directly improves its inference platform's competitive moat — multi-GPU kernel generation [P7](https://www.together.ai/blog/parallelkernelbench)[P9](https://github.com/togethercomputer/ParallelKernelBench), RL/post-training co-design [P8](https://job-boards.greenhouse.io/togetherai/jobs/5168929007), speculative decoding [E28](https://job-boards.greenhouse.io/togetherai/jobs/4987660007), and agent infrastructure [P20](https://job-boards.greenhouse.io/togetherai/jobs/5157380007). A new Head of Hyperscaler Partnerships role [P1](https://job-boards.greenhouse.io/togetherai/jobs/5171124007) signals intent to scale distribution through cloud platform deals, while the Amsterdam engineering hub is expanding as a dedicated center for identity, collaboration, and sandbox backend systems [P2](https://job-boards.greenhouse.io/togetherai/jobs/5169470007)[P13](https://job-boards.greenhouse.io/togetherai/jobs/5162910007)[P18](https://job-boards.greenhouse.io/togetherai/jobs/5068767007)[E32](https://job-boards.greenhouse.io/togetherai/jobs/5028862007). The research internship pipeline for Fall 2026 is unusually broad — spanning inference, model shaping, RL/post-training, GPU programming, and frontier agents — indicating sustained R&D investment across the full stack [E9](https://job-boards.greenhouse.io/togetherai/jobs/5157661007)[E10](https://job-boards.greenhouse.io/togetherai/jobs/5157759007)[E11](https://job-boards.greenhouse.io/togetherai/jobs/5157380007)[E15](https://job-boards.greenhouse.io/togetherai/jobs/5168929007)[E29](https://job-boards.greenhouse.io/togetherai/jobs/5157559007).\n\n## Signal desks\n\n### Hiring\n\n- **Identity & Collaboration team (Amsterdam):** Hiring both a Software Engineer and Senior Software Engineer to build authentication flows (SSO, OAuth, SAML), organizations, API keys, and role-based access controls in Elixir/Phoenix + TypeScript/Next.js. Signals enterprise tenant readiness and multi-user collaboration product surface [P2](https://job-boards.greenhouse.io/togetherai/jobs/5169470007)[P13](https://job-boards.greenhouse.io/togetherai/jobs/5162910007)[E4](https://job-boards.greenhouse.io/togetherai/jobs/5169470007)[E8](https://job-boards.greenhouse.io/togetherai/jobs/5162910007).\n- **Cloud infrastructure (San Francisco & Amsterdam):** Multiple roles — Senior Software Engineer – Together Cloud Infrastructure (SF and Amsterdam), AI Infrastructure Engineer, Staff Engineer for Distributed Storage and HPC & AI Infrastructure, Infrastructure Design Engineer — point to a major buildout of GPU clusters, managed storage, and networking [P3](https://job-boards.greenhouse.io/togetherai/jobs/5172169007)[E6](https://job-boards.greenhouse.io/togetherai/jobs/5172169007)[E22](https://job-boards.greenhouse.io/togetherai/jobs/5138540007)[E23](https://job-boards.greenhouse.io/togetherai/jobs/5155722007)[E30](https://job-boards.greenhouse.io/togetherai/jobs/4749787007)[E32](https://job-boards.greenhouse.io/togetherai/jobs/5028862007)[E35](https://job-boards.greenhouse.io/togetherai/jobs/5135876007).\n- **Product Manager, AI Infrastructure (San Francisco):** New role focused on GPU Clusters, Managed Storage, and observability, with a path to full product-area ownership within nine months. Indicates rapid surface expansion and formalization of infrastructure PM discipline [P3](https://job-boards.greenhouse.io/togetherai/jobs/5172169007)[E6](https://job-boards.greenhouse.io/togetherai/jobs/5172169007).\n- **Hyperscaler partnerships (San Francisco):** A principal-level Head of Hyperscaler Partnerships role reporting to the VP of Strategic Partnerships, with direct exposure to CEO/CFO/CRO. Explicitly covers model licensing, revshare, marketplace private offers, and scaled cloud distribution [P1](https://job-boards.greenhouse.io/togetherai/jobs/5171124007)[E2](https://job-boards.greenhouse.io/togetherai/jobs/5171124007).\n- **Research interns (Fall 2026, San Francisco):** Five distinct intern tracks — Inference [P19](https://job-boards.greenhouse.io/togetherai/jobs/5157759007)[E10](https://job-boards.greenhouse.io/togetherai/jobs/5157759007), Model Shaping [P17](https://job-boards.greenhouse.io/togetherai/jobs/5157661007)[E9](https://job-boards.greenhouse.io/togetherai/jobs/5157661007), RL & Post-Training Systems (Turbo) [P8](https://job-boards.greenhouse.io/togetherai/jobs/5168929007)[E15](https://job-boards.greenhouse.io/togetherai/jobs/5168929007), Frontier Agents [P20](https://job-boards.greenhouse.io/togetherai/jobs/5157380007)[E11](https://job-boards.greenhouse.io/togetherai/jobs/5157380007)[W1](https://zerogtalent.com/ai-jobs/together-ai/frontier-agents-intern-fall-2026-5157380007)[W2](https://underprompt.com/jobs/frontier-agents-intern-together-ai), and GPU Programming [P21](https://job-boards.greenhouse.io/togetherai/jobs/5157559007)[E29](https://job-boards.greenhouse.io/togetherai/jobs/5157559007). Pattern: all require empirical research rigor, Python + deep learning frameworks, and familiarity with Transformer architectures; several explicitly mention CUDA/C++ and systems co-design.\n- **Inference engineering (San Francisco/NYC):** Research Engineer for Frontier Speculative Decoding [E28](https://job-boards.greenhouse.io/togetherai/jobs/4987660007) and Machine Learning Engineer – Inference [E27](https://job-boards.greenhouse.io/togetherai/jobs/4385540007) signal continued heavy investment in inference speed as a differentiator.\n- **Data Center Operations (San Francisco):** Data Center Operations Coordinator role to manage break/fix across multiple data center locations, indicating self-operated or closely managed colocation footprint [P22](https://job-boards.greenhouse.io/togetherai/jobs/5160139007).\n- **Data platform & analytics:** Backend Software Engineer – Data Platform & AI Data Products [E37](https://job-boards.greenhouse.io/togetherai/jobs/5064263007), Analytics Engineer – Data Warehouse [E33](https://job-boards.greenhouse.io/togetherai/jobs/5101651007), Finance Analytics Engineer [E25](https://job-boards.greenhouse.io/togetherai/jobs/5108385007), and Infrastructure Accounting Manager [E26](https://job-boards.greenhouse.io/togetherai/jobs/5134279007) together suggest internal data infrastructure buildout and increasing financial/operational sophistication around GPU capacity economics.\n- **Customer support functions:** Customer Support Engineer roles for both Inference and GPU Cluster [E38](https://job-boards.greenhouse.io/togetherai/jobs/5147747007)[E40](https://job-boards.greenhouse.io/togetherai/jobs/5093510007) indicate growing paying customer base requiring dedicated production support.\n- **Voice AI:** Staff Machine Learning Engineer, Voice AI role referenced externally [W4](https://tryjeremy.com/jobs/together-ai-staff-machine-learning-engineer-voice-ai-019e4191ad74) aligns with Together's speech-to-text and TTS product investments [E50](https://www.together.ai/blog/how-together-ai-built-the-worlds-fastest-speech-to-text-stack)[E55](https://www.together.ai/blog/introducing-voice-finder-a-new-tool-to-quickly-find-the-right-voice-for-your-app-from-over-600-voices).\n- **Recruiting scale-up:** Senior Technical Recruiter for AI/ML Research [E21](https://job-boards.greenhouse.io/togetherai/jobs/5135941007) signals competitive research hiring push, consistent with the breadth of Fall 2026 intern roles.\n- **Workplace/operations:** Workplace Coordinator [P11](https://job-boards.greenhouse.io/togetherai/jobs/5166459007)[E16](https://job-boards.greenhouse.io/togetherai/jobs/5166459007) and Lead Product Designer [E39](https://job-boards.greenhouse.io/togetherai/jobs/5062829007) round out organizational scaling.\n\n### Forks\n\n- **InferenceX (fork of SemiAnalysisAI/InferenceX):** Apache-2.0 continuous inference benchmark platform targeting B200, GB200 NVL72, MI355X, GB300 NVL72, and TPU/ Trainium. Description explicitly mentions Kimi K2.6, DeepSeekv4, GLM5. Forked to run official/unofficial Together inference benchmarking [P23](https://github.com/togethercomputer/InferenceX)[E43](https://github.com/togethercomputer/InferenceX).\n- **DeepGEMM (fork of deepseek-ai/DeepGEMM):** Forked April 2026. Signals interest in DeepSeek's FP8 GEMM library, likely for inference kernel optimization on Together's serving stack [E60](https://github.com/togethercomputer/DeepGEMM).\n- **k8s-netperf (fork of cloud-bulldozer/k8s-netperf):** Kubernetes network performance benchmarking. Relevant to multi-datacenter GPU cluster networking and interconnect performance measurement [E59](https://github.com/togethercomputer/k8s-netperf).\n- **tinker-cookbook (fork of thinking-machines-lab/tinker-cookbook):** Forked June 2026. A cookbook for model training/ fine-tuning recipes; consistent with Model Shaping team's post-training and customization work [E46](https://github.com/togethercomputer/tinker-cookbook).\n- **Historical forks:** Port_FasterTransformer (NVIDIA/FasterTransformer) [P25](https://github.com/togethercomputer/Port_FasterTransformer), flash-attention (Dao-AILab) [P26](https://github.com/togethercomputer/flash-attention), diffusers (HazyResearch/diffusers) [P27](https://github.com/togethercomputer/diffusers), together-chat (AI-Yash/st-chat) [P28](https://github.com/togethercomputer/together-chat) — all from 2022–2023, inactive or archival, reflecting earlier-stage exploration.\n\n### Releases\n\n- **together-py SDK (v2.16.1 → v2.20.0):** Rapid iteration over ~10 days in late June 2026. New features: `whoami` CLI command and API endpoint [P4](https://github.com/togethercomputer/together-py/releases/tag/v2.18.0)[E7](https://github.com/togethercomputer/together-py/releases/tag/v2.18.0), CLI endpoint adapter commands [P10](https://github.com/togethercomputer/together-py/releases/tag/v2.17.0)[E14](https://github.com/togethercomputer/together-py/releases/tag/v2.17.0), remediation approval mode [P5](https://github.com/togethercomputer/together-py/releases/tag/v2.19.0)[E5](https://github.com/togethercomputer/together-py/releases/tag/v2.19.0), upload-path memory fix [P10](https://github.com/togethercomputer/together-py/releases/tag/v2.17.0). Pattern: CLI ergonomics, enterprise remediation workflows, and OpenAPI spec expansion. [E1](https://github.com/togethercomputer/together-py/releases/tag/v2.20.0)[E24](https://github.com/togethercomputer/together-py/releases/tag/v2.16.1)\n- **together-typescript SDK (v0.41.2):** Chore-heavy release adding staging CI syncing and remediation summary docs improvements [P14](https://github.com/togethercomputer/together-typescript/releases/tag/v0.41.2)[E19](https://github.com/togethercomputer/together-typescript/releases/tag/v0.41.2).\n- **together-sandbox (workspace v1.11.0 → v3.0.0):** Breaking change in v3.0.0 — `snapshots.list()` now returns cursor-paginated results instead of plain arrays [P12](https://github.com/togethercomputer/together-sandbox/releases/tag/together-sandbox-workspace-v3.0.0)[E18](https://github.com/togethercomputer/together-sandbox/releases/tag/together-sandbox-workspace-v3.0.0). Consistent iteration cadence throughout May–June 2026 [E45](https://github.com/togethercomputer/together-sandbox/releases/tag/together-sandbox-workspace-v2.0.0)[E49](https://github.com/togethercomputer/together-sandbox/releases/tag/together-sandbox-workspace-v1.12.0)[E51](https://github.com/togethercomputer/together-sandbox/releases/tag/together-sandbox-workspace-v1.11.0).\n- **detect_agent (v0.3.0):** Release tagged May 2026; agent detection tooling [E48](https://github.com/togethercomputer/detect_agent/releases/tag/detect_agent-v0.3.0).\n- **xorl-wheels (tilelang 0.1.10):** Prebuilt CUDA 13.1 wheels for TileLang, a kernel DSL — relevant to GPU programming research [E47](https://github.com/togethercomputer/xorl-wheels/releases/tag/tilelang_0.1.10_cu131).\n\n### Talking\n\n- **ParallelKernelBench (June 2026):** Flagship research post. Together tested frontier LLMs on 87 real-world multi-GPU CUDA kernel problems. Best model solved <33% correctly; fewer than 25% of those beat the naive PyTorch+NCCL baseline. A few generated kernels beat any public implementation, including one for NVIDIA NeMo-RL's GRPO training loop. Positions Together as the authority on inference-critical kernel benchmarking [P7](https://www.together.ai/blog/parallelkernelbench)[P9](https://github.com/togethercomputer/ParallelKernelBench)[E13](https://www.together.ai/blog/parallelkernelbench).\n- **Kimi K2.7 Code vs Claude Fable 5 (June 2026):** Competitive cost/quality analysis. Kimi K2.7 Code produced landing pages at 94% lower cost than Claude Fable 5, within a few quality points. Narrative: open-source models are closing the gap on quality while being dramatically cheaper — reinforces Together's cost-leadership positioning [P15](https://www.together.ai/blog/kimi-k2-7-code-vs-claude-fable-5)[E20](https://www.together.ai/blog/kimi-k2-7-code-vs-claude-fable-5).\n- **ISO 27001:2022 certification (June 2026):** Enterprise trust milestone. Certification covers global platform ISMS, corporate HQ, and third-party data centers. Authored by Together's security leadership. Directly addresses enterprise procurement requirements [P24](https://www.together.ai/blog/iso-27001-2022-certification)[E44](https://www.together.ai/blog/iso-27001-2022-certification).\n- **Serving MiniMax-M3 (June 2026):** Technical deep-dive on efficient inference for a 1M-token-context multimodal model using KV-block-major sparse attention, paged MSA decode, and a Rust-based multimodal gateway [E42](https://www.together.ai/blog/serving-minimax-m3-for-efficient-inference-unlocking-1m-token-context-and-multimodality-without-regrets).\n- **Speech-to-text stack (May 2026):** Claims world's fastest ASR on Artificial Analysis by treating speech-to-text as a full-path systems problem rather than GPU inference alone [E50](https://www.together.ai/blog/how-together-ai-built-the-worlds-fastest-speech-to-text-stack).\n- **Coding agent benchmarks (May 2026):** Claims 31% more TPS than TensorRT-LLM, 2× better TTFT at saturation, and 76% lower cost than Claude Opus 4.6 for coding agent workloads [E52](https://www.together.ai/blog/coding-agent-benchmarks).\n- **DeepSeek-V4 serving (May 2026):** Explores million-token context as an inference systems problem on HGX B200 — compressed KV layouts, prefix caching, kernel maturity [E56](https://www.together.ai/blog/serving-deepseek-v4-why-million-token-context-is-an-inference-systems-problem).\n- **Parcae (April 2026):** Research on stable looped models — a 770M model matching 1.3B Transformer quality, with first scaling laws for looping [E58](https://www.together.ai/blog/parcae).\n- **Product/partnership announcements:** GPT Image 2 availability on Together [W3](https://www.linkedin.com/posts/togethercomputer_were-excited-to-announce-gpt-image-2-on-activity-7473774288823549952-PiCd), Pearl Research Labs partnership for discounted inference via Proof of Useful Work [E53](https://www.together.ai/blog/together-ai-partners-with-pearl-research-labs), Violin open-source video translation [E54](https://www.together.ai/blog/violin-open-source-translation-skill), Voice Finder tool for 600+ TTS voices [E55](https://www.together.ai/blog/introducing-voice-finder-a-new-tool-to-quickly-find-the-right-voice-for-your-app-from-over-600-voices), HuggingFace one-click deploy via Goose [E57](https://www.together.ai/blog/deploy-and-inference-any-model-from-huggingface).\n\n## Shipping\n\nTogether's shipping velocity in late June 2026 is concentrated on SDK maturity: four Python SDK releases (v2.16.1 through v2.20.0) across ~10 days added CLI ergonomics (`whoami`, endpoint adapters), enterprise remediation approval workflows, and upload stability fixes [P4](https://github.com/togethercomputer/together-py/releases/tag/v2.18.0)[P5](https://github.com/togethercomputer/together-py/releases/tag/v2.19.0)[P10](https://github.com/togethercomputer/together-py/releases/tag/v2.17.0)[P16](https://github.com/togethercomputer/together-py/releases/tag/v2.16.1). The TypeScript SDK saw a chore/infra release focused on CI syncing [P14](https://github.com/togethercomputer/together-typescript/releases/tag/v0.41.2). The together-sandbox product shipped a breaking pagination change (v3.0.0) signaling production API hardening for snapshot-heavy workflows [P12](https://github.com/togethercomputer/together-sandbox/releases/tag/together-sandbox-workspace-v3.0.0). No new models or model cards shipped in the evidence window; shipping activity is overwhelmingly platform and developer-experience oriented.\n\n## Research themes\n\n1. **Multi-GPU kernel generation and benchmarking:** ParallelKernelBench is the headline research artifact — an 87-problem benchmark and evaluation framework for LLM-generated multi-GPU CUDA kernels, directly relevant to inference serving performance [P7](https://www.together.ai/blog/parallelkernelbench)[P9](https://github.com/togethercomputer/ParallelKernelBench)[E13](https://www.together.ai/blog/parallelkernelbench).\n2. **Inference-aware RL and post-training:** The Turbo team is co-designing RL algorithms with inference systems, studying how speculative decoding, KV cache management, and partial rollouts affect learning dynamics in GRPO/RLHF/DPO-style methods [P8](https://job-boards.greenhouse.io/togetherai/jobs/5168929007)[E15](https://job-boards.greenhouse.io/togetherai/jobs/5168929007).\n3. **Model shaping and efficient training:** Research interns are working on advanced post-training (supervised, preference optimization, RL), distributed training improvements, and foundation model evaluation [P17](https://job-boards.greenhouse.io/togetherai/jobs/5157661007)[E9](https://job-boards.greenhouse.io/togetherai/jobs/5157661007). The team has recent publications at ICML 2026 and ICLR 2026 [P17](https://job-boards.greenhouse.io/togetherai/jobs/5157661007).\n4. **Frontier agents:** The Agents team investigates post-training for agentic behavior, self-learning and long-horizon reasoning, evaluation frameworks for open-ended tasks, and agent infrastructure at scale [P20](https://job-boards.greenhouse.io/togetherai/jobs/5157380007)[E11](https://job-boards.greenhouse.io/togetherai/jobs/5157380007)[W1](https://zerogtalent.com/ai-jobs/together-ai/frontier-agents-intern-fall-2026-5157380007)[W2](https://underprompt.com/jobs/frontier-agents-intern-together-ai).\n5. **Speculative decoding and inference optimization:** Dedicated research engineer role for frontier speculative decoding [E28](https://job-boards.greenhouse.io/togetherai/jobs/4987660007), inference research intern covering KV cache design, compiler-aware optimization, and Mixture-of-Experts serving [P19](https://job-boards.greenhouse.io/togetherai/jobs/5157759007)[E10](https://job-boards.greenhouse.io/togetherai/jobs/5157759007).\n6. **GPU kernel co-design:** GPU Programming interns work on CUDA/Triton kernel optimization co-designed with model architecture, reflecting the hardware-software co-design philosophy articulated in Together's mission [P21](https://job-boards.greenhouse.io/togetherai/jobs/5157559007)[E29](https://job-boards.greenhouse.io/togetherai/jobs/5157559007)[E31](https://job-boards.greenhouse.io/togetherai/jobs/4188119007).\n7. **Looped/recurrent models:** Parcae research (April 2026) demonstrated stable looped LMs matching Transformers 2× their size, with scaling laws for recurrence [E58](https://www.together.ai/blog/parcae).\n\n## Hiring & scaling\n\nTogether AI is hiring across three geographic hubs — San Francisco (primary), Amsterdam (secondary engineering hub), and New York City (limited). The Amsterdam office is concentrated on Identity & Collaboration and Sandbox backend engineering [P2](https://job-boards.greenhouse.io/togetherai/jobs/5169470007)[P13](https://job-boards.greenhouse.io/togetherai/jobs/5162910007)[P18](https://job-boards.greenhouse.io/togetherai/jobs/5068767007)[E32](https://job-boards.greenhouse.io/togetherai/jobs/5028862007), while San Francisco carries the full breadth: cloud infrastructure, research, product, partnerships, data platform, and operations. The Fall 2026 intern cohort (five distinct tracks, 12–16 weeks, September–December) is substantial and spans the full research-to-systems spectrum [P8](https://job-boards.greenhouse.io/togetherai/jobs/5168929007)[P17](https://job-boards.greenhouse.io/togetherai/jobs/5157661007)[P19](https://job-boards.greenhouse.io/togetherai/jobs/5157759007)[P20](https://job-boards.greenhouse.io/togetherai/jobs/5157380007)[P21](https://job-boards.greenhouse.io/togetherai/jobs/5157559007). A dedicated Senior Technical Recruiter for AI/ML Research [E21](https://job-boards.greenhouse.io/togetherai/jobs/5135941007) and a Finance Analytics Engineer [E25](https://job-boards.greenhouse.io/togetherai/jobs/5108385007) plus Infrastructure Accounting Manager [E26](https://job-boards.greenhouse.io/togetherai/jobs/5134279007) suggest organizational scaling to support larger research headcount and increasingly complex GPU capacity economics. Data center operations hiring [P22](https://job-boards.greenhouse.io/togetherai/jobs/5160139007) indicates self-managed or closely-managed colocation rather than purely cloud-based infrastructure. The hyperscaler partnerships role [P1](https://job-boards.greenhouse.io/togetherai/jobs/5171124007) is a board-level strategic hire, suggesting major cloud distribution deals are in negotiation or planning.\n\n## Category implications\n\n- **Infrastructure:** Together is building a full-stack AI cloud — GPU clusters, managed storage, networking, and observability — not just an inference API [P3](https://job-boards.greenhouse.io/togetherai/jobs/5172169007)[E23](https://job-boards.greenhouse.io/togetherai/jobs/5155722007)[E35](https://job-boards.greenhouse.io/togetherai/jobs/5135876007). The Data Center Operations Coordinator [P22](https://job-boards.greenhouse.io/togetherai/jobs/5160139007) and Infrastructure Design Engineer [E35](https://job-boards.greenhouse.io/togetherai/jobs/5135876007) roles indicate physical infrastructure ownership. The InferenceX fork [P23](https://github.com/togethercomputer/InferenceX)[E43](https://github.com/togethercomputer/InferenceX) and k8s-netperf fork [E59](https://github.com/togethercomputer/k8s-netperf) signal systematic benchmarking across hardware targets (B200, GB200, MI355X, TPU). This positions Together as competing with both hyperscalers and neoclouds on infrastructure depth, not just model serving. The hyperscaler partnerships hire [P1](https://job-boards.greenhouse.io/togetherai/jobs/5171124007) suggests a co-opetition strategy: build your own infra while also distributing through the majors.\n- **Product:** The product surface is expanding rapidly from inference into storage, observability, sandbox environments, and enterprise identity/access management. The together-sandbox workspace v3.0.0 with pagination [P12](https://github.com/togethercomputer/together-sandbox/releases/tag/together-sandbox-workspace-v3.0.0) and the Identity & Collaboration team buildout [P2](https://job-boards.greenhouse.io/togetherai/jobs/5169470007)[P13](https://job-boards.greenhouse.io/togetherai/jobs/5162910007) point to a product evolving from developer tool to enterprise platform. Customer support engineers for both inference and GPU clusters [E38](https://job-boards.greenhouse.io/togetherai/jobs/5147747007)[E40](https://job-boards.greenhouse.io/togetherai/jobs/5093510007) confirm paying production customers.\n- **Research:** Together's research strategy is tightly coupled to its infrastructure business — kernel benchmarking [P7](https://www.together.ai/blog/parallelkernelbench), speculative decoding [E28](https://job-boards.greenhouse.io/togetherai/jobs/4987660007), inference-aware RL [P8](https://job-boards.greenhouse.io/togetherai/jobs/5168929007), and GPU programming [P21](https://job-boards.greenhouse.io/togetherai/jobs/5157559007) all directly improve inference cost and speed, which is the company's core GTM differentiator. The Model Shaping team's post-training work [P17](https://job-boards.greenhouse.io/togetherai/jobs/5157661007)[E9](https://job-boards.greenhouse.io/togetherai/jobs/5157661007) extends this into the customization layer, enabling customers to fine-tune on Together's infrastructure. Frontier agents research [P20](https://job-boards.greenhouse.io/togetherai/jobs/5157380007) and Parcae looped models [E58](https://www.together.ai/blog/parcae) represent longer-horizon bets.\n- **GTM:** ISO 27001:2022 certification [P24](https://www.together.ai/blog/iso-27001-2022-certification)[E44](https://www.together.ai/blog/iso-27001-2022-certification) is a deliberate enterprise GTM unlock. The cost-comparison blog posts (Kimi vs. Claude at 94% less cost [P15](https://www.together.ai/blog/kimi-k2-7-code-vs-claude-fable-5), coding agents at 76% lower cost than Opus [E52](https://www.together.ai/blog/coding-agent-benchmarks)) are direct price-performance marketing aimed at developers and engineering teams. The Pearl Research Labs partnership [E53](https://www.together.ai/blog/together-ai-partners-with-pearl-research-labs) explores novel distribution via crypto-economic channels. The Head of Hyperscaler Partnerships [P1](https://job-boards.greenhouse.io/togetherai/jobs/5171124007) signals intent to pursue marketplace and revshare deals with major cloud platforms as a complementary GTM motion.\n- **Strategy:** The evidence reveals a three-layer strategy: (1) be the fastest and cheapest inference platform through deep systems research and kernel optimization, (2) expand into adjacent infrastructure (storage, networking, observability, sandboxes) to increase platform stickiness and revenue per customer, (3) pursue enterprise adoption through compliance (ISO 27001), identity/access features, and hyperscaler marketplace distribution. The breadth of Fall 2026 research interns — spanning inference, post-training, agents, and GPU programming — suggests Together is betting that maintaining inference performance leadership requires sustained, multi-disciplinary research investment rather than one-off optimizations.\n\n## Traction highlights\n\n- SDK release velocity: four together-py releases in ~10 days (v2.16.1 through v2.20.0) with feature-bearing changes [P4](https://github.com/togethercomputer/together-py/releases/tag/v2.18.0)[P5](https://github.com/togethercomputer/together-py/releases/tag/v2.19.0)[P10](https://github.com/togethercomputer/together-py/releases/tag/v2.17.0)[P16](https://github.com/togethercomputer/together-py/releases/tag/v2.16.1)[E1](https://github.com/togethercomputer/together-py/releases/tag/v2.20.0)[E7](https://github.com/togethercomputer/together-py/releases/tag/v2.18.0)[E5](https://github.com/togethercomputer/together-py/releases/tag/v2.19.0)[E14](https://github.com/togethercomputer/together-py/releases/tag/v2.17.0)[E24](https://github.com/togethercomputer/together-py/releases/tag/v2.16.1).\n- together-sandbox: three workspace releases in May–June 2026 (v1.11.0, v1.12.0, v2.0.0, v3.0.0) with breaking API changes indicating production usage [P12](https://github.com/togethercomputer/together-sandbox/releases/tag/together-sandbox-workspace-v3.0.0)[E18](https://github.com/togethercomputer/together-sandbox/releases/tag/together-sandbox-workspace-v3.0.0)[E45](https://github.com/togethercomputer/together-sandbox/releases/tag/together-sandbox-workspace-v2.0.0)[E49](https://github.com/togethercomputer/together-sandbox/releases/tag/together-sandbox-workspace-v1.12.0)[E51](https://github.com/togethercomputer/together-sandbox/releases/tag/together-sandbox-workspace-v1.11.0).\n- Enterprise certification: ISO 27001:2022 achieved via ANAB-accredited A-LIGN audit, covering global platform and third-party data centers [P24](https://www.together.ai/blog/iso-27001-2022-certification)[E44](https://www.together.ai/blog/iso-27001-2022-certification).\n- Research output: ParallelKernelBench paper with code, HuggingFace dataset, and public blog post [P7](https://www.together.ai/blog/parallelkernelbench)[P9](https://github.com/togethercomputer/ParallelKernelBench)[E13](https://www.together.ai/blog/parallelkernelbench); Parcae paper on looped models [E58](https://www.together.ai/blog/parcae); ICML 2026 and ICLR 2026 publications from Model Shaping interns [P17](https://job-boards.greenhouse.io/togetherai/jobs/5157661007).\n- Competitive benchmarking claims: 31% more TPS than TensorRT-LLM for coding agents [E52](https://www.together.ai/blog/coding-agent-benchmarks), world's fastest speech-to-text on Artificial Analysis [E50](https://www.together.ai/blog/how-together-ai-built-the-worlds-fastest-speech-to-text-stack), 94% cost reduction vs. Claude Fable 5 for landing page generation [P15](https://www.together.ai/blog/kimi-k2-7-code-vs-claude-fable-5)[E20](https://www.together.ai/blog/kimi-k2-7-code-vs-claude-fable-5).\n- Model serving breadth: MiniMax-M3 (1M context, multimodal) [E42](https://www.together.ai/blog/serving-minimax-m3-for-efficient-inference-unlocking-1m-token-context-and-multimodality-without-regrets), DeepSeek-V4 (1M context) [E56](https://www.together.ai/blog/serving-deepseek-v4-why-million-token-context-is-an-inference-systems-problem), GPT Image 2 [W3](https://www.linkedin.com/posts/togethercomputer_were-excited-to-announce-gpt-image-2-on-activity-7473774288823549952-PiCd), Kimi K2.7 Code [P15](https://www.together.ai/blog/kimi-k2-7-code-vs-claude-fable-5), Gemma-4-31B-it-pearl [E53](https://www.together.ai/blog/together-ai-partners-with-pearl-research-labs), HuggingFace any-model deployment [E57](https://www.together.ai/blog/deploy-and-inference-any-model-from-huggingface).\n- GitHub organization activity: 8+ active repositories with recent commits including ParallelKernelBench [P9](https://github.com/togethercomputer/ParallelKernelBench), InferenceX fork [P23](https://github.com/togethercomputer/InferenceX), together-storage-claude-skills (Go-based runbooks for T4+CS3 storage) [E3](https://github.com/togethercomputer/together-storage-claude-skills), together-sandbox, together-py, together-typescript, detect_agent, and xorl-wheels.\n\n## Sources\n\n- [P1](https://job-boards.greenhouse.io/togetherai/jobs/5171124007) Head of Hyperscaler Partnerships job listing\n- [P2](https://job-boards.greenhouse.io/togetherai/jobs/5169470007) Software Engineer (Amsterdam) job listing\n- [P3](https://job-boards.greenhouse.io/togetherai/jobs/5172169007) Product Manager, AI Infrastructure job listing\n- [P4](https://github.com/togethercomputer/together-py/releases/tag/v2.18.0) together-py v2.18.0 release\n- [P5](https://github.com/togethercomputer/together-py/releases/tag/v2.19.0) together-py v2.19.0 release\n- [P6](https://job-boards.greenhouse.io/togetherai/jobs/4790243007) Platform Engineer, Model Shaping job listing\n- [P7](https://www.together.ai/blog/parallelkernelbench) ParallelKernelBench blog post\n- [P8](https://job-boards.greenhouse.io/togetherai/jobs/5168929007) Research Intern RL & Post-Training Systems, Turbo job listing\n- [P9](https://github.com/togethercomputer/ParallelKernelBench) ParallelKernelBench repository metadata\n- [P10](https://github.com/togethercomputer/together-py/releases/tag/v2.17.0) together-py v2.17.0 release\n- [P11](https://job-boards.greenhouse.io/togetherai/jobs/5166459007) Workplace Coordinator job listing\n- [P12](https://github.com/togethercomputer/together-sandbox/releases/tag/together-sandbox-workspace-v3.0.0) together-sandbox workspace v3.0.0 release\n- [P13](https://job-boards.greenhouse.io/togetherai/jobs/5162910007) Senior Software Engineer (Amsterdam) job listing\n- [P14](https://github.com/togethercomputer/together-typescript/releases/tag/v0.41.2) together-typescript v0.41.2 release\n- [P15](https://www.together.ai/blog/kimi-k2-7-code-vs-claude-fable-5) Kimi K2.7 Code vs Claude Fable 5 blog post\n- [P16](https://github.com/togethercomputer/together-py/releases/tag/v2.16.1) together-py v2.16.1 release\n- [P17](https://job-boards.greenhouse.io/togetherai/jobs/5157661007) Research Intern, Model Shaping (Fall 2026) job listing\n- [P18](https://job-boards.greenhouse.io/togetherai/jobs/5068767007) Backend Engineer job listing\n- [P19](https://job-boards.greenhouse.io/togetherai/jobs/5157759007) Research Intern, Inference (Fall 2026) job listing\n- [P20](https://job-boards.greenhouse.io/togetherai/jobs/5157380007) Frontier Agents Intern (Fall 2026) job listing\n- [P21](https://job-boards.greenhouse.io/togetherai/jobs/5157559007) Systems Research Engineer Intern - GPU Programming job listing\n- [P22](https://job-boards.greenhouse.io/togetherai/jobs/5160139007) Data Center Operations Coordinator job listing\n- [P23](https://github.com/togethercomputer/InferenceX) InferenceX repository metadata\n- [P24](https://www.together.ai/blog/iso-27001-2022-certification) ISO 27001:2022 certification blog post\n- [P25](https://github.com/togethercomputer/Port_FasterTransformer) Port_FasterTransformer repository metadata\n- [P26](https://github.com/togethercomputer/flash-attention) flash-attention repository metadata\n- [P27](https://github.com/togethercomputer/diffusers) diffusers repository metadata\n- [P28](https://github.com/togethercomputer/together-chat) together-chat repository metadata\n- Event stream entries as cited throughout\n- [W1](https://zerogtalent.com/ai-jobs/together-ai/frontier-agents-intern-fall-2026-5157380007) Frontier Agents Intern - Zero G Talent\n- [W2](https://underprompt.com/jobs/frontier-agents-intern-together-ai) Frontier Agents Intern - Underprompt\n- [W3](https://www.linkedin.com/posts/togethercomputer_were-excited-to-announce-gpt-image-2-on-activity-7473774288823549952-PiCd) GPT Image 2 LinkedIn announcement\n- [W4](https://tryjeremy.com/jobs/together-ai-staff-machine-learning-engineer-voice-ai-019e4191ad74) Staff ML Engineer, Voice AI - Jeremy","generated_at":"2026-06-27T18:44:11.988+00:00","citations":[{"url":"https://job-boards.greenhouse.io/togetherai/jobs/5172169007","path":null,"label":"job-boards.greenhouse.io/togetherai","type":"external"},{"url":"https://job-boards.greenhouse.io/togetherai/jobs/5155722007","path":null,"label":"job-boards.greenhouse.io/togetherai","type":"external"},{"url":"https://www.together.ai/blog/iso-27001-2022-certification","path":null,"label":"together.ai/blog","type":"external"},{"url":"https://job-boards.greenhouse.io/togetherai/jobs/5169470007","path":null,"label":"job-boards.greenhouse.io/togetherai","type":"external"},{"url":"https://job-boards.greenhouse.io/togetherai/jobs/5162910007","path":null,"label":"job-boards.greenhouse.io/togetherai","type":"external"},{"url":"https://www.together.ai/blog/parallelkernelbench","path":null,"label":"together.ai/blog","type":"external"},{"url":"https://github.com/togethercomputer/ParallelKernelBench","path":null,"label":"togethercomputer/ParallelKernelBench","type":"external"},{"url":"https://job-boards.greenhouse.io/togetherai/jobs/5168929007","path":null,"label":"job-boards.greenhouse.io/togetherai","type":"external"},{"url":"https://job-boards.greenhouse.io/togetherai/jobs/4987660007","path":null,"label":"job-boards.greenhouse.io/togetherai","type":"external"},{"url":"https://job-boards.greenhouse.io/togetherai/jobs/5157380007","path":null,"label":"job-boards.greenhouse.io/togetherai","type":"external"},{"url":"https://job-boards.greenhouse.io/togetherai/jobs/5171124007","path":null,"label":"job-boards.greenhouse.io/togetherai","type":"external"},{"url":"https://job-boards.greenhouse.io/togetherai/jobs/5068767007","path":null,"label":"job-boards.greenhouse.io/togetherai","type":"external"},{"url":"https://job-boards.greenhouse.io/togetherai/jobs/5028862007","path":null,"label":"job-boards.greenhouse.io/togetherai","type":"external"},{"url":"https://job-boards.greenhouse.io/togetherai/jobs/5157661007","path":null,"label":"job-boards.greenhouse.io/togetherai","type":"external"},{"url":"https://job-boards.greenhouse.io/togetherai/jobs/5157759007","path":null,"label":"job-boards.greenhouse.io/togetherai","type":"external"},{"url":"https://job-boards.greenhouse.io/togetherai/jobs/5157380007","path":null,"label":"job-boards.greenhouse.io/togetherai","type":"external"},{"url":"https://job-boards.greenhouse.io/togetherai/jobs/5168929007","path":null,"label":"job-boards.greenhouse.io/togetherai","type":"external"},{"url":"https://job-boards.greenhouse.io/togetherai/jobs/5157559007","path":null,"label":"job-boards.greenhouse.io/togetherai","type":"external"},{"url":"https://job-boards.greenhouse.io/togetherai/jobs/5169470007","path":null,"label":"job-boards.greenhouse.io/togetherai","type":"external"},{"url":"https://job-boards.greenhouse.io/togetherai/jobs/5162910007","path":null,"label":"job-boards.greenhouse.io/togetherai","type":"external"},{"url":"https://job-boards.greenhouse.io/togetherai/jobs/5172169007","path":null,"label":"job-boards.greenhouse.io/togetherai","type":"external"},{"url":"https://job-boards.greenhouse.io/togetherai/jobs/5138540007","path":null,"label":"job-boards.greenhouse.io/togetherai","type":"external"},{"url":"https://job-boards.greenhouse.io/togetherai/jobs/4749787007","path":null,"label":"job-boards.greenhouse.io/togetherai","type":"external"},{"url":"https://job-boards.greenhouse.io/togetherai/jobs/5135876007","path":null,"label":"job-boards.greenhouse.io/togetherai","type":"external"},{"url":"https://job-boards.greenhouse.io/togetherai/jobs/5171124007","path":null,"label":"job-boards.greenhouse.io/togetherai","type":"external"},{"url":"https://job-boards.greenhouse.io/togetherai/jobs/5157759007","path":null,"label":"job-boards.greenhouse.io/togetherai","type":"external"},{"url":"https://job-boards.greenhouse.io/togetherai/jobs/5157661007","path":null,"label":"job-boards.greenhouse.io/togetherai","type":"external"},{"url":"https://zerogtalent.com/ai-jobs/together-ai/frontier-agents-intern-fall-2026-5157380007","path":null,"label":"zerogtalent.com/ai-jobs","type":"external"},{"url":"https://underprompt.com/jobs/frontier-agents-intern-together-ai","path":null,"label":"underprompt.com/jobs","type":"external"},{"url":"https://job-boards.greenhouse.io/togetherai/jobs/5157559007","path":null,"label":"job-boards.greenhouse.io/togetherai","type":"external"},{"url":"https://job-boards.greenhouse.io/togetherai/jobs/4385540007","path":null,"label":"job-boards.greenhouse.io/togetherai","type":"external"},{"url":"https://job-boards.greenhouse.io/togetherai/jobs/5160139007","path":null,"label":"job-boards.greenhouse.io/togetherai","type":"external"},{"url":"https://job-boards.greenhouse.io/togetherai/jobs/5064263007","path":null,"label":"job-boards.greenhouse.io/togetherai","type":"external"},{"url":"https://job-boards.greenhouse.io/togetherai/jobs/5101651007","path":null,"label":"job-boards.greenhouse.io/togetherai","type":"external"},{"url":"https://job-boards.greenhouse.io/togetherai/jobs/5108385007","path":null,"label":"job-boards.greenhouse.io/togetherai","type":"external"},{"url":"https://job-boards.greenhouse.io/togetherai/jobs/5134279007","path":null,"label":"job-boards.greenhouse.io/togetherai","type":"external"},{"url":"https://job-boards.greenhouse.io/togetherai/jobs/5147747007","path":null,"label":"job-boards.greenhouse.io/togetherai","type":"external"},{"url":"https://job-boards.greenhouse.io/togetherai/jobs/5093510007","path":null,"label":"job-boards.greenhouse.io/togetherai","type":"external"},{"url":"https://tryjeremy.com/jobs/together-ai-staff-machine-learning-engineer-voice-ai-019e4191ad74","path":null,"label":"tryjeremy.com/jobs","type":"external"},{"url":"https://www.together.ai/blog/how-together-ai-built-the-worlds-fastest-speech-to-text-stack","path":null,"label":"together.ai/blog","type":"external"},{"url":"https://www.together.ai/blog/introducing-voice-finder-a-new-tool-to-quickly-find-the-right-voice-for-your-app-from-over-600-voices","path":null,"label":"together.ai/blog","type":"external"},{"url":"https://job-boards.greenhouse.io/togetherai/jobs/5135941007","path":null,"label":"job-boards.greenhouse.io/togetherai","type":"external"},{"url":"https://job-boards.greenhouse.io/togetherai/jobs/5166459007","path":null,"label":"job-boards.greenhouse.io/togetherai","type":"external"},{"url":"https://job-boards.greenhouse.io/togetherai/jobs/5166459007","path":null,"label":"job-boards.greenhouse.io/togetherai","type":"external"},{"url":"https://job-boards.greenhouse.io/togetherai/jobs/5062829007","path":null,"label":"job-boards.greenhouse.io/togetherai","type":"external"},{"url":"https://github.com/togethercomputer/InferenceX","path":null,"label":"togethercomputer/InferenceX","type":"external"},{"url":"https://github.com/togethercomputer/InferenceX","path":null,"label":"togethercomputer/InferenceX","type":"external"},{"url":"https://github.com/togethercomputer/DeepGEMM","path":null,"label":"togethercomputer/DeepGEMM","type":"external"},{"url":"https://github.com/togethercomputer/k8s-netperf","path":null,"label":"togethercomputer/k8s-netperf","type":"external"},{"url":"https://github.com/togethercomputer/tinker-cookbook","path":null,"label":"togethercomputer/tinker-cookbook","type":"external"},{"url":"https://github.com/togethercomputer/Port_FasterTransformer","path":null,"label":"togethercomputer/Port_FasterTransformer","type":"external"},{"url":"https://github.com/togethercomputer/flash-attention","path":null,"label":"togethercomputer/flash-attention","type":"external"},{"url":"https://github.com/togethercomputer/diffusers","path":null,"label":"togethercomputer/diffusers","type":"external"},{"url":"https://github.com/togethercomputer/together-chat","path":null,"label":"togethercomputer/together-chat","type":"external"},{"url":"https://github.com/togethercomputer/together-py/releases/tag/v2.18.0","path":null,"label":"togethercomputer/together-py","type":"external"},{"url":"https://github.com/togethercomputer/together-py/releases/tag/v2.18.0","path":null,"label":"togethercomputer/together-py","type":"external"},{"url":"https://github.com/togethercomputer/together-py/releases/tag/v2.17.0","path":null,"label":"togethercomputer/together-py","type":"external"},{"url":"https://github.com/togethercomputer/together-py/releases/tag/v2.17.0","path":null,"label":"togethercomputer/together-py","type":"external"},{"url":"https://github.com/togethercomputer/together-py/releases/tag/v2.19.0","path":null,"label":"togethercomputer/together-py","type":"external"},{"url":"https://github.com/togethercomputer/together-py/releases/tag/v2.19.0","path":null,"label":"togethercomputer/together-py","type":"external"},{"url":"https://github.com/togethercomputer/together-py/releases/tag/v2.20.0","path":null,"label":"togethercomputer/together-py","type":"external"},{"url":"https://github.com/togethercomputer/together-py/releases/tag/v2.16.1","path":null,"label":"togethercomputer/together-py","type":"external"},{"url":"https://github.com/togethercomputer/together-typescript/releases/tag/v0.41.2","path":null,"label":"togethercomputer/together-typescript","type":"external"},{"url":"https://github.com/togethercomputer/together-typescript/releases/tag/v0.41.2","path":null,"label":"togethercomputer/together-typescript","type":"external"},{"url":"https://github.com/togethercomputer/together-sandbox/releases/tag/together-sandbox-workspace-v3.0.0","path":null,"label":"togethercomputer/together-sandbox","type":"external"},{"url":"https://github.com/togethercomputer/together-sandbox/releases/tag/together-sandbox-workspace-v3.0.0","path":null,"label":"togethercomputer/together-sandbox","type":"external"},{"url":"https://github.com/togethercomputer/together-sandbox/releases/tag/together-sandbox-workspace-v2.0.0","path":null,"label":"togethercomputer/together-sandbox","type":"external"},{"url":"https://github.com/togethercomputer/together-sandbox/releases/tag/together-sandbox-workspace-v1.12.0","path":null,"label":"togethercomputer/together-sandbox","type":"external"},{"url":"https://github.com/togethercomputer/together-sandbox/releases/tag/together-sandbox-workspace-v1.11.0","path":null,"label":"togethercomputer/together-sandbox","type":"external"},{"url":"https://github.com/togethercomputer/detect_agent/releases/tag/detect_agent-v0.3.0","path":null,"label":"togethercomputer/detect_agent","type":"external"},{"url":"https://github.com/togethercomputer/xorl-wheels/releases/tag/tilelang_0.1.10_cu131","path":null,"label":"togethercomputer/xorl-wheels","type":"external"},{"url":"https://www.together.ai/blog/parallelkernelbench","path":null,"label":"together.ai/blog","type":"external"},{"url":"https://www.together.ai/blog/kimi-k2-7-code-vs-claude-fable-5","path":null,"label":"together.ai/blog","type":"external"},{"url":"https://www.together.ai/blog/kimi-k2-7-code-vs-claude-fable-5","path":null,"label":"together.ai/blog","type":"external"},{"url":"https://www.together.ai/blog/iso-27001-2022-certification","path":null,"label":"together.ai/blog","type":"external"},{"url":"https://www.together.ai/blog/serving-minimax-m3-for-efficient-inference-unlocking-1m-token-context-and-multimodality-without-regrets","path":null,"label":"together.ai/blog","type":"external"},{"url":"https://www.together.ai/blog/coding-agent-benchmarks","path":null,"label":"together.ai/blog","type":"external"},{"url":"https://www.together.ai/blog/serving-deepseek-v4-why-million-token-context-is-an-inference-systems-problem","path":null,"label":"together.ai/blog","type":"external"},{"url":"https://www.together.ai/blog/parcae","path":null,"label":"together.ai/blog","type":"external"},{"url":"https://www.linkedin.com/posts/togethercomputer_were-excited-to-announce-gpt-image-2-on-activity-7473774288823549952-PiCd","path":null,"label":"linkedin.com/posts","type":"external"},{"url":"https://www.together.ai/blog/together-ai-partners-with-pearl-research-labs","path":null,"label":"together.ai/blog","type":"external"},{"url":"https://www.together.ai/blog/violin-open-source-translation-skill","path":null,"label":"together.ai/blog","type":"external"},{"url":"https://www.together.ai/blog/deploy-and-inference-any-model-from-huggingface","path":null,"label":"together.ai/blog","type":"external"},{"url":"https://github.com/togethercomputer/together-py/releases/tag/v2.16.1","path":null,"label":"togethercomputer/together-py","type":"external"},{"url":"https://job-boards.greenhouse.io/togetherai/jobs/4188119007","path":null,"label":"job-boards.greenhouse.io/togetherai","type":"external"},{"url":"https://github.com/togethercomputer/together-storage-claude-skills","path":null,"label":"togethercomputer/together-storage-claude-skills","type":"external"},{"url":"https://job-boards.greenhouse.io/togetherai/jobs/4790243007","path":null,"label":"job-boards.greenhouse.io/togetherai","type":"external"}],"provenance":{"provider":"deepseek","model":"deepseek-v4-pro","workflow":"onlylabs-deepagents-analysis-v3","agent":"deepagents"},"evidence":{"total":92,"pages":28,"events":140,"web":4,"signal_desks":{"forks":4,"repos":1,"hiring":31,"talking":12,"releases":12},"data_radar_lanes":null,"data_radar_matches":null}},"signal_counts":{"total":515,"model_released":30,"release":212,"repo_new":41,"repo_forked":59,"post_published":103,"job_opened":70}}