Thought leadership on data engineering and AI. Agentic data engineering, programming languages for data, model training, and the case studies we run on our own platform.
Stop gambling on one agent. vibeCodeCLI races parallel fleets in your terminal, with persistent memory across sessions and outcome-aware safety gates that stop dangerous commands before they execute. Free. 23 providers. Bring your own keys.
We rebuilt ThinkingMemory around one primitive: recall. Intent in, the right context out, hybrid-retrieved, ranked, and packed to a token budget, with an in-engine lifecycle, bitemporal audit, per-tenant RLS, and an MCP server. 100% recall@5 in roughly 68% fewer tokens than dumping the corpus.
GPT-4-class capability cost $37.50/M tokens in 2023. Today a 2B-parameter model serves the same benchmark score at $0.04. The frontier held its price; the efficient tier collapsed. The same open model spans a 13× price spread across 12 providers. Built on Bonacci Studio.
Four public datasets (BLS/OEWS, O*NET, Anthropic Economic Index, Indeed). Three occupation coding schemes. One crosswalk. AI has not cut jobs in aggregate, but it is bending the hiring curve along the automation line before headcount moves. Built in Bonacci Studio.
25 months of activity from GitHub, arXiv, and Hugging Face, 613 data points across five categories. RAG is up 340% year over year. Training from scratch is up 32% month over month. Nothing is cooling. The agent wrote the PySpark, ran it, and rendered the analysis live in Bonacci Studio.
A new research paper introducing AC-LSCM, an Action-Conditioned Latent Structural Causal Model that gives agents an explicit forward model with interventional semantics. 36× fewer safety violations than a Transformer baseline on synthetic planning tasks.
How I built three 9B Mixture-of-Experts models trained on 100B tokens each for under $10K using Google's TPU Research Cloud program. Domain-specific SLMs that excel at their tasks while keeping costs low.
Meet the autonomous AI agent powered by ThinkingMemory that lives inside Bonacci Studio. It remembers context, executes SQL, SSH, APIs, and MCP autonomously, monitors your infrastructure 24/7, and routes alerts to Slack, Discord, Teams, Email, PagerDuty, and more. Use TL or PySpark scripts as custom heartbeat checks.
Data deserves its own language. ThinkingLanguage (TL) is a compiled, statically-typed programming language where tables, streams, tensors, and models are native types, not library add-ons. Replace Python, SQL, Spark, and YAML with one fast, safe language.
A unified workspace with Terminal, AI Chat, and Codegen tabs where AI agents with real tool access autonomously explore databases, write ETL pipelines, and execute PySpark jobs through natural language.
Our open-source memory infrastructure for AI agents, with layered architecture and reasoning-aware retrieval. Make agents cheaper, more consistent, and genuinely improving over time.
Our AI-powered platform delivering concise tech and AI news summaries in 60 words or less. Stay informed efficiently in today's fast-paced technology landscape.
Bonacci Studio's AI feature that automatically detects, analyzes, and resolves data structure changes to prevent silent data corruption in your pipelines.
Bonacci Studio's robust Kafka Connect ecosystem with 16+ connectors, providing mature CDC capabilities and bi-directional data flow for event-driven architectures.
Why ThinkingDBx chose self-managed Linux VPS over cloud providers, saving 95% in server costs while maintaining control and flexibility.
The case studies above are real pipelines: the agent wrote the code, ran it on Spark, and rendered the results. Describe yours and watch it ship.