Carlos D. Escobar-Valbuena

Agent OS architect and AI engineering lead. I build AI-native, multi-tenant data platforms in production — across regulated and high-stakes domains. Concurrent senior leadership across Stimulus (AI Lead), TEAM International (Data Architect), and Wedi Pay (Co-founder/CTO, 2024–2026). Open-source author of Life Agent OS, the bstack agentic harness + broomva/skills, Lago, Vigil, and the RCS paper series. MSc AI at Universidad de los Andes (2026).

Live signals

Pulled directly from GitHub and crates.io — what the substrate is actually doing this week.

GitHub stars
0
across 0 repos
crates.io downloads
0
0 Life Agent OS crates published
Recent writing
0
posts + notes in latest cycle
Most recent push
2h ago
broomva.tech

Updated hourly · 79 ★ · 1.4k crate downloads

GitHub activity

The same plots on my GitHub profile — each card is a live SVG that re-renders from the GitHub API, so it stays current on its own.

7+yrs shipping AI platforms140public repos14Hugging Face models
Overall stats & rank
GitHub stats — total stars, commits, PRs, issues, and rank
Top languages
Most-used languages across public repositories
Contribution activity
Contribution activity over the past weeks
Contribution streak
Current and longest GitHub contribution streak

Knowledge graph

The Broomva bookkeeping pipeline scores and promotes raw extracts into a queryable entity graph. This is what's in the graph right now — last sync 3mo ago.

Entities
164
in the graph
Top-scored
49
scored 8/9 or 9/9
Last 7 days
80
new promotions
Composition
concept · 64pattern · 35project · 29tool · 17person · 12question · 6

Concurrent leadership

Three Principal-tier roles in parallel since late 2024 — compressed surface area across fintech / payments, agentic procurement, and enterprise data platforms.

Nov 2024 – Present

AI Lead

Stimulus

Architecture redesign of an AI-native, Databricks-backed agentic procurement platform serving FIFA 2026 and the Olympics Committee. Lakehouse-native real-time substrate: Databricks Zerobus streaming into a Unity-Catalog-governed medallion architecture (bronze/silver/gold). Multi-document RAG with policy-bounded tool adapters; multi-tenant JWT hardening.

Oct 2024 – Jun 2026

Co-founder & CTO

Wedi Pay

Agentic cross-border B2B payment orchestration. Multi-tenant Postgres RLS, Kafka/Redpanda event-driven backbone, Databricks gold-table ETLs. Wedi Agents execute payment actions under strict policy/consent/scoped credentials with full audit trails. Signed and shipped payment-rail integrations — Prometeo (open-banking A2A), TruBit/VelaFi (USDC on/off-ramp), Thirdweb (web3-native) — on a provider-agnostic integration layer.

Nov 2020 – Present

Data Architect (Contract)

TEAM International

Architectural oversight on the Time Series Engine pipelines I previously led as Sr. ML/AI Tech Lead. Anchor customer: a NYSE-listed US natural-gas operator with ~70K wells across multiple US basins. Multi-tenant streaming + batch telemetry on Databricks + MLflow + PyTorch handling 10K+ records/sec. Production medallion-style bronze/silver/gold tables since 2021.

Selected engagements

Anonymized references — case studies available under mutual NDA.

LATAM payments integration layer
3 payment rails · multi-country

Signed and shipped integrations with Prometeo (open-banking account-to-account), TruBit — now VelaFi (USDC on/off-ramp), and Thirdweb (web3-native) behind one provider-agnostic layer at Wedi Pay.

NYSE-listed US natural-gas operator
~70,000 wells · multi-basin

Anchor customer of the Time Series Engine. Production medallion-style lakehouse since 2021 with anomaly detection, alarm threshold estimation, and digital-twin integration.

Tier-1 sports + procurement enterprise
FIFA 2026 + Olympics Committee scale

Stimulus customer surface — agentic supplier intelligence with policy-bounded tool adapters and multi-document retrieval.

US-based AI advisory
AWS Premier Consulting Partner

Production-grade agentic RAG over an internal SageMaker engineering knowledge base — pgvector + FAISS, three switchable interaction modes (graph-based, tools agent, conversational). Delivery accepted at evaluation stage.

Recommendation carousel ranking
Tier-1 LATAM e-commerce + fintech

Polars-based ML pipeline integrating prints/taps/payments event streams into a feature-engineered training set; GridSearchCV tuning with full MLflow autolog reproducibility.

Open source & research

The substrate behind everything else — published as MIT/Apache OSS.

Life Agent OS

Rust agent runtime kernel. Multi-tenant gateway with TLS 1.3, JWKS, KMS abstraction (Vault Transit / AWS KMS / GCP KMS), token-bucket rate limiting, event-sourced persistence on redb v2, OpenTelemetry-native observability. 15+ crates on crates.io.

Lago

Event-sourced persistence substrate. Append-only event journal on redb v2; full provenance for every state transition.

Vigil

OpenTelemetry-native observability foundation for the Life Agent OS. Single crate, four modules, traces + metrics + logs unified.

Haima

Agentic finance engine. x402 metered machine-to-machine payments with USDC settlement on Base. Six Rust crates; secp256k1 wallet, per-task billing, full audit trails.

bstack + broomva/skills

Agentic development harness — 20 irreducible automation primitives (policy gates, knowledge bookkeeping, CI productive-wait, persistent cross-context loops, cross-model adversarial review) that turn an agent-driven workspace into a self-operating system, plus 48 portable agent skills (agentskills.io spec) for Claude Code, Codex, Cursor, Gemini CLI, Goose, and Copilot.

Recursive Controlled Systems (RCS)

Five-paper series formalizing LLM-as-controller agents as a 7-tuple Σ = (X, Y, U, f, h, S, Π) with recursive stability budgets across hierarchical control levels (L0–L3). Control-native architecture for autonomous workflows — verification causally independent of the acting model — and the ongoing autonomy / search / intelligence research program. Foundations paper complete; 4 papers in progress.

Most-starred repos

Live from GitHub. Updated hourly.

harness-engineering

25

Harness Engineering playbook — deterministic smoke/test/lint harness commands, strict architecture boundaries, entropy-control checks for autonomous agent workflows.

Shell·1mo ago

symphony

9

Rust orchestration daemon for coding agents — polls Linear/GitHub/Markdown, creates isolated workspaces, dispatches agent sessions with EGRI self-improvement.

Rust·19d ago

vortex

4

Versatile Orchestrated Execution Engine for Data & AI Pipelines | LLM-powered Smart Agent for Workflows Development and Automation

Python·1y ago

skills

3

Broomva agent-skills monorepo — 48 Tier-2 skills compatible with Claude Code, Codex, Cursor, Gemini CLI, Goose, Copilot. Layout follows anthropics/skills (agentskills.io spec). Install: npx skills add broomva/skills --skill <name>.

Python·3d ago

aiOS

3

Kernel contract for the Life Agent OS — canonical types, runtime ports, event taxonomy, policy gates. The genome that all subsystems implement.

Rust·21d ago

arcan

3

Core agent runtime — event loop, LLM providers (Anthropic, OpenAI), capability system, TUI. Part of the Life Agent OS.

Rust·9d ago

Life Agent OS — published crates

Live from crates.io. The Rust substrate of the agent runtime.

life-vigil

v0.3.0

Observability foundation for the Life Agent OS — OpenTelemetry tracing, GenAI semantic conventions, and contract-derived instrumentation

689 downloads3mo ago

life-aios

v0.3.0

Kernel contract facade for the Life Agent OS — re-exports aios-protocol, aios-events, aios-runtime, and aios-kernel

115 downloads3mo ago

life-lago

v0.3.0

Persistence substrate facade for the Life Agent OS — re-exports lago-core, lago-journal, lago-store, lago-fs, lago-api, and lago-policy

114 downloads3mo ago

life-praxis

v0.3.0

Tool execution engine facade for the Life Agent OS — re-exports praxis-core, praxis-tools, praxis-skills, and praxis-mcp-bridge

101 downloads3mo ago

life-haima

v0.3.0

Agentic finance engine facade for the Life Agent OS — re-exports haima-core, haima-wallet, haima-x402, and haima-api

101 downloads3mo ago

life-autonomic

v0.3.0

Homeostasis controller facade for the Life Agent OS — re-exports autonomic-core, autonomic-controller, and autonomic-api

101 downloads3mo ago

life-anima

v0.3.0

Identity and beliefs facade for the Life Agent OS — re-exports anima-core, anima-identity, and anima-lago

100 downloads3mo ago

life-arcan

v0.3.0

Agent daemon and runtime facade for the Life Agent OS — re-exports arcan-core, arcand, arcan-harness, arcan-store, arcan-provider, and arcan-aios-adapters

68 downloads3mo ago

Recent writing

Long-form essays and short notes from the build.

Writing

The Benchmark Ran Out Before the Loop Did

We pre-registered a self-improvement experiment, and it was over in one generation — the model aced every task and the selection gate became mathematically unsatisfiable. Two weeks later another lab hit the same wall. The instrument didn't fail. The instrument class did.

July 21, 2026

The Folklore Arrived Before the Physics

In six months the field converged on a belief — the harness, not the model, is what matters — and backed it with striking empirics. What it doesn't have is a single theorem about when the loop it now worships converges, oscillates, or collapses. A map of the gap, and the four results that would fill it.

July 21, 2026

Forking Houston: 89 commits, six agents, one hack camp

We took Houston — an MIT-licensed agent runtime — and grafted a new brain onto it. In a single hack camp the fork went 89 commits ahead of upstream: four net-new Rust crates, a full Linear integration, a feature-flag platform, a push pipeline. The features are the easy part to show. The real story is the method that produced them — autonomous agents working in parallel git worktrees under a governance harness that made the speed safe instead of reckless.

June 6, 2026

Notes

Autonomy is governance

The degree of autonomy you can grant an agent is directly proportional to the quality of your governance infrastructure.

March 14, 2026

Harness over prompting

Prompt quality matters, but harness quality is what keeps agent outputs reliable once stakes increase.

February 17, 2026

A release rhythm that works

One weekly long-form post plus short daily notes creates enough surface area for learning, distribution, and iteration.

February 14, 2026

Stack

Lakehouse & Data Platforms

Databricks (deep)Unity Catalog RBACZerobus streamingMedallion architectureDelta LakeSnowflakePySparkIcebergTrinoKafka / RedpandaDagsterMLflow

AI / RAG / Agentic

Anthropic ClaudeOpenAIAWS BedrockLangChainLangGraphLangSmithLangfuseMCPpgvectorFAISSBM25 hybrid retrievalLLM-as-judge

Cloud & Observability

AWSAzureGCPKubernetesTerraformOpenTelemetry (vigil author)CloudWatchDatadog

Languages & Runtime

PythonRustTypeScriptSQLNext.jsFastAPIPyTorchHugging Face

Elsewhere

GitHub

@Broomva

OSS substrate, agent runtime, control metalayer, and the bstack skill catalog.

LinkedIn

in/broomva

Career timeline, role progression, and connections.

X / Twitter

@broomva_tech

Notes from the build — agent OS, RCS papers, and what I'm thinking about.

Download CV

Three lengths — pick the one that fits your read time.

5 pp

Master CV

Full surface — every role, OSS, research, education.

Download PDF
3 pp

Targeted (AI Data Architect)

Lakehouse + RAG + governance focus.

Download PDF
1 pp

Executive

One-page top-line summary.

Download PDF

Talk to me

If you're building production agent systems, multi-tenant lakehouses, or governed AI platforms — and want to compare architectures or discuss a role — open the contact page with your specific bottleneck. Detailed engagement narratives available under mutual NDA.

Contact optionsWhat I'm building now

broomva.tech

Reliability engineering for complex systems.

  • Pages
  • Home
  • Projects
  • Writing
  • Notes
  • Tools
  • Chat
  • Prompts
  • Link Hub
  • Social
  • GitHub
  • LinkedIn
  • X