Engagements that ship. Senior IC. Fixed scope.

Five fixed-scope engagements, one monthly fractional seat. Real-time platforms. Production RAG. Agents that hold under live traffic. No junior hand-offs.

Currently accepting 2 fixed-scope engagements + 1 fractional seat for Q3. EU AI Act high-risk enforcement starts Dec 2027. Sprint seats are limited.

Book a 45-minute scoping call

No pitch deck. If I can't move the needle on your ROI in 30 minutes, I'll tell you for free what would.

Six ways to work together

Five fixed-scope engagements. One monthly retainer. Senior IC, every engagement. No hourly billing.

New · Front-door offer

14-Day RAG MVP Sprint · $9,500

Show your board what production RAG looks like. In 14 days. Ships a working RAG pipeline on a slice of your data, an evaluation harness with a 30–50 query golden set, and a cost baseline. Not a demo — a deployable prototype. Self-liquidating: most teams upgrade to the full RAG engagement. IP-free handoff on the final invoice.

Kick off your sprint
01
Premium

RAG Pipeline Engineering

Cut inference costs 52% without sacrificing retrieval quality.

Caching, routing, quantization, hybrid search, and eval harness — applied to your corpus, your traffic patterns, your SLAs. Delivered with a runbook your team can own on Day 31.

You get: production RAG pipeline + eval harness + cost baseline + runbook

Deliverables

  • Domain-tuned chunking & embedding strategy
  • Hybrid retrieval (BM25 + dense) + cross-encoder reranking
  • Evaluation harness with golden-set regression tests in CI
  • Langfuse / Phoenix tracing wired into your stack
  • Handoff runbook + recorded architecture walkthrough
FromTimeline
$25,0004–8 weeks
Scope your pipeline
02
Premium

Agentic Workflow Orchestration

Agents that complete tasks. Not agents that need hand-holding.

Multi-agent orchestration with human-in-the-loop checkpoints, observability dashboards, and failure-mode recovery. Designed for production traffic, not LangChain tutorials.

You get: multi-agent system + observability dashboard + failure-mode recovery + runbook

Deliverables

  • Agent topology design + failure-mode analysis
  • Tool-use contracts with strict schema validation
  • Per-step cost & latency budgets enforced at runtime
  • Human-in-the-loop escalation paths
  • Observability dashboard (traces, tokens, retries, alerts)
FromTimeline
$50,0006–10 weeks
Design your agent system
03
Premium

AI Architecture Audit

Find the failure modes before your users do.

Production readiness, security posture, cost efficiency, and EU AI Act gap analysis. You get a 1-page executive summary + a prioritized remediation roadmap. 2 weeks. $15K. One page of findings. Or nothing.

You get: 1-page exec summary + prioritized remediation roadmap + 30-day support

Deliverables

  • Written audit (40–60 pages) with severity-graded findings
  • Executive summary (1 page) for leadership
  • Prioritized remediation roadmap (P0 / P1 / P2)
  • 60-minute live readout with Q&A
  • 30 days of async follow-up in a shared channel
  • Risk-framework mapping (EU AI Act · NIST AI RMF · SOC2) when in scope
FromTimeline
$15,0002 weeks
Book your audit

Deliverable preview · AI Compliance Readiness Sprint

Executive Summary · 1 page

Anonymized. Format shown reflects the structure of every delivered engagement.

1 · Risk classification

System: career-coach-assistant-v3 · EU AI Act

  • High-risk (Annex III §4) · employment / career-coaching decisions affecting access to work
  • NIST AI RMF · Govern + Map partially complete, Measure + Manage in flight
  • SOC 2 Type II · 3 controls fail (CC6.1, CC7.2, CC8.1: logging, monitoring, change mgmt)

2 · Top 3 findings (P0)

  1. F1 No human-review checkpoint on adverse decisions. Required by Art. 14.
  2. F2 Training-data lineage untraceable. GDPR Art. 22 + EU AI Act Art. 10 conflict.
  3. F3 No bias / disparate-impact testing on demographic-protected classes.

3 · Remediation roadmap

P0 Human-in-the-loop checkpoint (1 sprint)
P1 Data lineage tool + audit log wiring (3 sprints)
P2 Bias eval harness in CI (2 sprints)

Bottom line

3 P0 fixes before your auditor finds them.

Estimated 6-8 sprints of work. Q3 enforcement starts Dec 2027. You have runway.

Numbers, vendors, and system names are illustrative. The structure (risk classification + 3 P0 findings + 3-tier roadmap + bottom line) is the actual deliverable shape.

05
Premium

AI Compliance Readiness Sprint

EU AI Act. NIST AI RMF. SOC 2. Audit-ready in 3 weeks.

Risk classification, training-data lineage, bias testing, human-review checkpoints, and documentation — delivered as an auditor-ready package. Enforcement window: Dec 2027. You have runway. Don't waste it.

You get: risk classification + training-data lineage + bias testing + auditor-ready docs

Deliverables

  • Risk classification (EU AI Act tier, NIST AI RMF function, SOC2 control set)
  • AI Bill of Materials: model provenance, training data lineage, dependencies, hardware config, eval suite (per Scharinger's AIBOM framework)
  • Model Card + Dataset Card + Tech Documentation Pack ready for your auditor
  • Gap analysis: what your auditor will catch and how to fix it before they do
  • Bias / fairness / robustness evaluation harness wired to your CI
  • 30-60 page readiness report with severity-graded findings + P0/P1/P2 roadmap
  • 1-page executive summary for legal / GRC / leadership
  • 60-minute readout with named compliance counsel (optional)
FromTimeline
$25,0003 weeks
Start your sprint
04
Premium

Fractional AI Architect

A senior AI architect, in your corner. 10 hours a week.

Embedded principal for teams scaling their AI platform. Senior judgment on architecture, code review, RFC drafting, vendor selection, and unblocking your engineers — without a $500K CAIO hire. Best fit when you're shipping v2 with constraints, need a senior to mentor a junior hire, or want an interim AI lead before a full-time exec is justified. 90-day minimum. Capacity-led, not sales-led.

Deliverables

  • 10 hrs/week of senior AI architecture time
  • Weekly written status + 60-min live sync
  • Architecture decision records (ADRs) + RFC review
  • Vendor / model / infrastructure evaluation
  • Async Slack response within 24 hours
  • Mentorship for your internal AI engineers
  • IP-free: no claim on your codebase
FromTimeline
$12,000/moMonthly · 90-day minimum
Apply for the seatRead the full engagement structure
06
Premium

14-Day RAG MVP Sprint

Show your board what production RAG looks like. In 14 days.

Your data. Your use case. A working RAG pipeline with eval harness and cost baseline. Not a demo — a deployable prototype. Self-liquidating: most teams upgrade to the full RAG engagement.

Deliverables

  • Working RAG pipeline on a slice of your data (not the full corpus)
  • Evaluation harness with 30–50 query golden set
  • 1-page diagnostic: where your current system fails + 3 prioritized fixes
  • Langfuse / Phoenix tracing wired into the prototype
  • 14 days of async support post-delivery
FromTimeline
$9,5002 weeks
Kick off your sprint

How I work

Three predictable phases. No surprise invoices. No scope creep. Every engagement runs the same playbook.

  1. Week 1

    Discover

    Stakeholder interviews, system audit, success metrics, and a written scoping document we both sign before any code ships.

  2. Weeks 2–3

    Design

    Architecture diagrams, evaluation harness, and a working prototype against your real data. Fail fast. Iterate in the open.

  3. Week 4+

    Deliver

    Production deployment, observability dashboards wired up, handoff runbook, and async support during the first 30 days.

Anti-promises

What I refuse.

Six lines that protect the work. If any of these disqualifies you, this site just saved us both an hour.

  1. 01

    I refuse hourly billing.

    Fixed scope or monthly retainer. Always.

  2. 02

    I refuse engagements under $15k — the audit is the floor.

    Below the Architecture Audit, I can't deliver principal-tier depth.

  3. 03

    I refuse more than 2 engagements per quarter.

    Quality compounds with attention, not headcount.

  4. 04

    I refuse ndas before a paid call.

    If the scoping call surfaces confidential IP, I'll sign yours.

  5. 05

    I refuse junior hand-offs.

    I take your engagement, I deliver every line.

  6. 06

    I refuse rebrand work, consumer apps, crypto, ad-tech.

    Real-time AI, RAG, agents, edge platforms. That's the work.

If any of this ruled you out, we saved each other an hour. If it didn't — book the audit.

Common questions

If yours isn't here, ask on the call.

How long does an engagement last?

Most engagements run 4–10 weeks depending on scope. The audit is a fixed 2 weeks. The RAG pipeline is typically 4–8, and orchestration 6–10.

What's the payment structure?

50% upfront on signed scope, 50% on delivery for fixed-scope engagements. The audit is split 50/50 with Net-15 on the second half. Fractional retainers bill monthly, 90-day minimum. I invoice in USD; wire or Stripe accepted.

Who owns the IP?

You do. Every line of code, every prompt, every eval dataset transfers to your repo on the final invoice. I'm happy to sign your NDA before the first call.

What time zones do you overlap?

Based in IST (UTC+5:30). Async-friendly by default; live syncs land comfortably across US, EU, and APAC mornings. Weekly written status + 30-min sync on your schedule.

Can you work with my existing team?

Yes. Most engagements embed with one of your engineers for the duration. I leave behind runbooks, recorded walkthroughs, and 30 days of async support.

AI Architecture Audit vs. AI Compliance Readiness Sprint: which fits?

Pick the AI Architecture Audit if your AI is in production and you need a senior read on inference, retrieval, cost, security, and team velocity. Pick the Compliance Readiness Sprint if your AI is shipping into the EU market, touches regulated data, or your auditor / regulator is on a clock. Many teams do both, six to ten weeks apart.

Or: ship the BI underneath

Need the dashboard, not just the AI?

Same senior hand. Shipped a Looker suite for JerseySTEM (US 501(c)(3)) that turned raw Salesforce data into the one number the ED, the board, and the auditor all read. RAG pipelines clean the input, LookML governs the metric, Looker surfaces it. If your AI project is dying because the underlying BI isn't trusted, this is the same engagement from the dashboard side.

See the JerseySTEM case study

Not sure which one fits?

Book a 45-minute scoping call. If I can't move the needle on your ROI in 30 minutes, I'll tell you for free what would.

Book a 45-minute consultation