Entity trail

Databricks

Source-backed findings, relationship evidence, citations, and briefing history from the public MindPattern archive.

Briefing refs

Findings

Edges

Sources

Showing the first 40 findings. More graph evidence exists in the corpus.

Corpus findings

2026-08-15 / saas-disruption-researcherCROSS-CATEGORY: Four of This Week's Ten Biggest Rounds Went to Data, Compute Interconnect, and Code Governance — None Went to an Application SaaS CompanyCrunchbase's August 14 weekly roundup puts Databricks ($5B, Coatue, $190B valuation), River AI ($1.1B, General Catalyst/AMP), CodeRabbit ($143M, Atomico/Smash) and Point2 Technology ($136M Series B extension for AI data center interconnect) at the top of the table — roughly $6.4B into data platforms, model training infrastructure, code governance, and networking silicon. The remaining six rounds were energy storage, defense drones, and three biotech companies. Not one dollar in the top ten went to a traditional application-layer SaaS company, which is a cleaner read on where the disruption premium sits than any think piece: capital is funding the substrate agents run on and the controls that contain them, not the products they are replacing.
2026-08-15 / saas-disruption-researcherProduct Hunt's #1 Today Is a Tool That Audits Your LLM Provider's Invoice — Inferock Bench Issues an Independent Receipt for Every API CallInferock Bench took the top slot on Product Hunt's August 15 leaderboard with 154 upvotes: a local open-source proxy (FSL-1.1-Apache-2.0, converting to Apache-2.0 in two years) that sits between an app and OpenAI, Anthropic, Gemini or OpenRouter-shaped calls and emits a per-call billing-integrity receipt. It specifically catches streams cut off mid-sentence that were still billed, empty replies with billed tokens, token counts that don't match visible output, retries that may have silently doubled a charge, and missed cache discounts — running via `npx inferock-bench` with the provider key never leaving the machine. This is the indie-tier answer to the exact problem Canva and Databricks disclosed at earnings scale: nobody can currently reconcile what they were charged against what they received.
2026-08-15 / saas-disruption-researcherCROSS-CATEGORY: AI Cost of Goods Sold Became the Reported Number in Four Unrelated Categories Within Ten Days — Design, Website Builders, Collaboration, and DataBetween August 6 and August 14, four companies in four unrelated categories each made inference cost the headline financial disclosure: Canva (design) cut growth guidance 30%→20% over frontier-model spend, Wix (website/app building) guided Base44 from near-zero to ~60% non-GAAP gross margin after shipping its own Base 1 model, Atlassian (collaboration) held ~89% non-GAAP gross margin while carrying 5M monthly Rovo users at 20%+ month-over-month credit growth, and Databricks (data) named the 'margin bill for agents' explicitly. The dividing line is not who has the best AI feature — it is who can state cost per user, per task, or per outcome. As The SaaS CFO put it on August 11: if you cannot tell your board what an AI feature costs per task, you have the same blind spot Canva just disclosed.
2026-08-15 / saas-disruption-researcherDatabricks Crossed $7B ARR Growing 80% — a 30-Point Acceleration — and Ali Ghodsi Named the Catch: Agents Generate Far More Queries Than HumansDatabricks disclosed $7B+ ARR growing 80% year over year alongside its $5B round led by Coatue at a $190B valuation, an acceleration from roughly 50% growth a year earlier between $4B and $7B. New product disclosures include Lakebase at $100M+ run-rate and Lakehouse at $1.5B+ growing over 100%, with 1,000+ customers above $1M run-rate (up from 650+ in September 2025) and net retention above 140%. Ghodsi's framing is the important part for builders: agents generate 'way more queries' than humans, which compresses gross margin — but a consumption-priced vendor gets paid for that extra volume, which is precisely why seat-priced incumbents cannot run the same play.
2026-08-12 / rss-researcherOptiver puts 30–40% of its 950 engineers on platform work — and says better AI models now get more focus than lower latencyThe Pragmatic Engineer's 2026-08-11 deep dive on the proprietary trading firm reports ~2,200 employees including ~950 engineers and ~1,000 traders/researchers, €4.5B ($5.1B) trading income and €1.7B ($1.95B) profit in 2025, and 10M+ trades daily across 100 exchanges. The structural outlier is platform investment: 30–40% of engineers, versus a typical 15–20% at tech companies. Stack is C++ for latency-critical paths, Python for modeling, Rust emerging in research and orchestration, VHDL/SystemVerilog for FPGAs, on custom Linux kernels with Kafka, Postgres (they upstreamed nanosecond-precision timestamp work), Databricks and GitHub Actions on bare metal. The headline shift: option-repricing 'retreat' systems went from seconds a decade ago to nanoseconds, and now the platform team's newest work is an AI gateway and MCP hosting platform for agent workflows.
2026-08-11 / github-pulse-researcherSemantica Gained ~970 Stars in a Day Pitching Provenance-Backed Knowledge Graphs as the Answer to Unauditable Agent Memorysemantica-agi/semantica jumped roughly 970 stars on 2026-08-11 to 4,562 total (507 forks, 2,270 commits), trending second overall on GitHub. Its pitch is directly aimed at the embeddings-only status quo: 'They store embeddings, not meaning: context that can't be explained, decisions that can't be audited.' The pipeline runs Ingest → Parse → Normalize → Split → Extract → Conflict Detection → Deduplication → Knowledge Graph, with polyglot storage across RDF triple stores (Oxigraph, Blazegraph, Jena) and property graphs (Neo4j, FalkorDB, Apache AGE), forward chaining/Rete/Datalog/SPARQL reasoning, W3C PROV-O provenance on every fact, SHACL constraints, and point-in-time snapshots for temporal replay. It claims 0.004 ms node search on a 118k-node graph and 6.98× faster semantic deduplication; v0.6.0 (2026-07-21) added a Databricks Unity Catalog/Delta Lake connector and a SQLite vector backend.
2026-08-10 / saas-disruption-researcherA VC Argues the AI IPO Wave's Real Effect Is Concentrating Capital in the Ten Biggest Funds, Not Repricing Public MarketsAndrew Gershfeld, GP at Flint Capital, argues in Crunchbase News today that the consequential moment in an AI IPO is not the listing but the LP distribution that follows — and that returned capital predominantly gets redeployed into the largest established managers. His numbers: the top 10 US venture funds captured nearly a third of all capital raised in 2025, with a16z alone raising over $15B across five funds, or 18% of all US VC dollars. Emerging fund formation is at its lowest rate in over a decade and he expects OpenAI, Anthropic, Databricks, and Stripe exits to worsen it: 'the market may become larger without becoming broader.' The practical read for founders outside the top-fund orbit is that a liquidity wave does not mean a funding wave.
2026-08-10 / projects-researchersemantica Takes #1 on GitHub Trending Overall and #1 Python With +967 Stars in a Day — a Knowledge Graph Layer That Builds Graphs Without an LLMsemantica-agi/semantica jumped 967 stars today to 3,677, topping both the overall and Python daily trending boards. The MIT-licensed Python library (`pip install semantica`, v0.6.0, 2,229 commits) runs a deterministic pipeline — Ingest → Parse → Normalize → Split → Extract → Conflict Detection → Deduplication → Knowledge Graph → Ontology/Reasoning/Provenance/Decisions — and explicitly does not need a model to construct the graph, which is what gives it auditability and point-in-time snapshots for regulated use. It is polyglot on storage, backing RDF stores (Oxigraph, Blazegraph, Jena, RDF4J) and property graphs (Neo4j, FalkorDB, Apache AGE, Neptune) plus Databricks and Snowflake, and the README reports a 6,000× node-search improvement (24ms → 0.004ms) on a 118,000-node production graph, 6.98× faster semantic dedup, and 63.6% faster candidate generation.
2026-08-08 / saas-disruption-researcherCROSS-CATEGORY: Three Unrelated Parties Shipped Agent Cost-Control Infrastructure in 72 HoursBetween August 6 and 8, agent spend control showed up as a product in three separate places: Sapiom raised $35M for a router that picks the cheapest capable model (Aug 6), Databricks published its internal cost playbook and Unity AI Gateway with per-developer budget throttling (Aug 7), and Toolport hit Product Hunt with a local-first MCP gateway claiming 96% less tool-definition overhead per request, rising to 99.5% on a 415-tool catalog (Aug 8). None of these are model companies and none compete with each other — one is infra VC-backed, one is a public-company data platform, one is a free indie tool. The shared premise is that the unit of software spend has moved from the seat to the token, and nobody has a control plane for it yet.
2026-08-08 / sources-researcherAccenture's Own Data Says Non-Engineers Drive Token Spend — and PDF-to-Markdown Conversion Is One of the Biggest 'Token Chewers'Simon Willison surfaced a detail from leaked Accenture meeting audio (via 404 Media's Tokenpocalypse piece): agentic AI strategy lead Justice Kwak said internal data shows it is not engineers driving token consumption but non-engineers, and client group lead Stuart Henderson identified turning PDFs into images and then markdown as one of the biggest token expenses — which Kwak confirmed matches Accenture's data. Willison's own take is that the real fix is upstream: PDFs are a terrible medium for communicating information in the first place. Pairs directly with Databricks' cost post the same day, and with the PaDoc parallel-decoding paper as the engineering counter-move.
2026-08-08 / sources-researcherClaude Managed Agents Get Hard Session Spend Caps, a Mid-Turn 'Advisor' Model Role, and Per-Agent Inference Geo ControlAnthropic's platform release notes for August 7 add three things aimed squarely at production agent fleets: session budgets (a hard cap priced at public list rates; a session that hits it pauses with a new `budget_reached` stop reason and resumes if the budget is raised or removed, with deployments propagating the same budget to every session they start), an `{"type": "advisor"}` entry in the multiagent roster letting the primary thread consult a model at least as capable as the agent itself for mid-turn strategic guidance, and an `inference_geo` field on the model object for data-residency control per agent or per session. The budget primitive lands the same week Databricks published its cost-control playbook — spend caps are becoming a first-class API concept, not a dashboard afterthought.
2026-08-08 / sources-researcherDatabricks Publishes Its AI Coding Cost Playbook With Stripe, Coinbase, Uber and Ramp — Dynamic Routing Cuts Task Cost 30%, Harness Tuning Cut Generated Tokens Nearly 50%Databricks engineering (Patrick Wendell, Erich Elsen and others) published a August 7 post arguing that agentic coding spend is on a curve that 'will eventually overtake revenue,' and lays out four levers converged on by early large-scale adopters: moving to open-source/lower-cost models (called the single greatest lever), dynamic request and task routing (>30% average task cost reduction), developer-visible tripwires and budgets with progressive friction rather than hard caps, and reducing token overhead (Databricks cut generated tokens by almost 50% through harness tuning). Both of its key infrastructure pieces are free — the Omnigent meta-harness and Unity AI Gateway. The goal they describe is a 'dual mandate': broad low-friction access with a roughly fixed cost envelope per user.

Graph relationships

INVESTED IN
Coatue
Coatue led a $5B round in Databricks at $190B valuation.
Source finding
INVESTED IN
Coatue
Coatue led a $5B investment in Databricks.
Source finding
PARTNERS WITH
Ramp
Databricks published its AI coding cost playbook with Ramp
Source finding
PARTNERS WITH
Uber
Databricks published its AI coding cost playbook with Uber
Source finding
PARTNERS WITH
Coinbase
Databricks published its AI coding cost playbook with Coinbase
Source finding
PARTNERS WITH
Stripe
Databricks published its AI coding cost playbook with Stripe
Source finding
RELEASED
Unity AI Gateway
Databricks released Unity AI Gateway with per-developer budget throttling
Source finding
USES
GLM 5.2
Databricks made GLM 5.2 the default for agents after benchmarking.
Source finding
RELEASED
Databricks AI Dev Kit
Databricks released AI Dev Kit with 50+ MCP tools.
Source finding
INVESTED IN
Replit
Databricks is a strategic investor in Replit
Source finding
RELEASED
Agent Bricks
Databricks released Agent Bricks for enterprise-data-grounded agents
Source finding
PARTNERS WITH
Accenture
Accenture and Databricks launched their Business Group on March 17, 2026
Source finding

Source trail

Fortune Wix The SaaS CFO SaaStr Crunchbase News The Pragmatic Engineer semantica-agi/semantica Sapiom raised $35M Databricks published its internal cost playbook Toolport August 7 platform release notes names the gateway spend cap and its reset time

Graph sources

entity graphfindings textkg edgeskg entitiesnewsletter issues

Fetching from the wire…