Insurance Operations · On-Chain Intelligence · AI Engineering
7+ years across subrogation, reinsurance, and commercial claims. Built tracking systems, recovered assets, handled $80M catastrophic losses.
Blockchain analytics, financial pattern recognition, and AI-powered tools — built independently since 2017.
7+ years recovering assets, preparing claims bordereaux, and building tools to make insurance operations work faster. $3M+ recovered. 300–400 concurrent claims. 5+ countries.
I've spent my career in claims operations — subrogation recovery, commercial claims handling, and reinsurance reporting across the UAE and the US. At Crawford, I ran a nationwide subrogation portfolio for 7-Eleven and recovered over $3 million. At RAK Insurance, I handled the claims side of inward reinsurance reporting — bordereaux, reserve validation, multi-currency reconciliation across international counterparties spanning UAE, Qatar, Saudi Arabia, Lebanon, and France.
I figure things out. When the claims system at Crawford was too slow, I built my own Excel tracking framework that pushed throughput up 30%. When I needed to process unstructured financial documents faster, I built an AI-powered application that turns a 6-hour due diligence questionnaire into 10 minutes of structured output with exact citations. I work independently, and I'd rather get the work right than talk about it.
I stepped away from traditional employment in 2020 for family caregiving. I didn't stop working — I spent the time building technical skills (Python, SQL, AI/LLM integrations), doing independent financial analysis across large datasets, detecting fraud patterns, and building production tools. Four years of daily analytical work that maps directly to claims operations. I'm ready to bring all of it back to insurance.
I don't just do one of these. I do all three — and that's what makes the difference.
End-to-end recovery from investigation through closure. Evidence packaging, demand development, negotiation, and settlement. Built custom tracking systems to manage high-volume portfolios.
Commercial claims across 6+ coverage lines with reinsurance claims reporting responsibilities. Monthly bordereaux preparation, reserve adequacy reviews, and multi-party reconciliation with counterparties across UAE, Qatar, Saudi Arabia, Lebanon, France, US, and Canada.
Built operational tools when existing systems weren't enough. Python, SQL, advanced Excel, and AI/LLM integrations applied to insurance workflows — document processing, claims triage, evidence review.
The numbers behind the work. All verified, all mine.
Started in the UAE, moved to the US, never stopped building.
Ran the 7-Eleven subrogation portfolio nationwide as a TPA — drive-thru hits, dispenser damage, signage, storefront impacts across the US and Canada.
1st and 3rd-party auto PD claims. Fast-moving volume — daily intake plus a large active queue. Had to balance closure speed with documentation sufficiency and claim defensibility.
Non-motor commercial claims from notification through to closure, plus sole responsibility for the claims side of reinsurance reporting on top of regular claims handling.
Stepped away for family caregiving. Didn't stop working — spent the time building technical skills and doing independent financial analysis.
I needed something, it didn't exist, so I built it.
I got tired of important information being buried in messy PDFs. So I built an app that takes unstructured insurance and financial documents, redacts PII on the browser side, runs them through LLMs, and gives back structured Excel outputs with exact citations — page, paragraph, line number. A DDQ that normally takes 5-6 hours? Done in under 10 minutes.
Set up a local AI that runs entirely on my machine — no data leaves the device. Handles document analysis, Q&A, and research tasks without touching external APIs. Built the full environment from scratch: model setup, Python config, troubleshooting. If you're dealing with PII or sensitive claims data, this is how you use AI without the risk.
Custom multi-sheet Excel framework built to manage 300–400 concurrent subrogation files at Crawford. Tracked demand status, response deadlines, recovery totals, missing documents, and bottleneck identification. Color-coded, status-based. Improved throughput by approximately 30%.
SQL-based reporting dashboards for monitoring large-scale transaction activity across multiple counterparties. Threshold-based alerting, capital flow tracking, and pattern detection.
Open to claims operations, subrogation, reinsurance, and InsurTech roles. Available for Dallas-based or remote positions.
I've tracked whale wallets, spotted manipulation setups from wallet activity alone, recovered $3M+ in assets through investigations, grew $5K to over $300K through on-chain analysis and disciplined risk management, and built production AI systems. I build what I need — Python, SQL, LLM APIs, whatever gets the job done.
I've been taking things apart to understand how they work since I was 8 — my dad's cassette radio, VCRs, CD players, early flat screens — whatever he brought home. Built my first PC from components at 15, no internet, no guides, just manuals and figuring it out. That same instinct is why I went on-chain in 2017 and never stopped.
Grew $5K of personal capital to over $300K through disciplined position management, on-chain analysis, and risk controls I built myself. I've tracked whale wallets, spotted manipulation setups from wallet activity alone, and analyzed fund flows across most major L1s and L2s. Deep hands-on experience across the full DeFi spectrum: BTC (spot, Ordinals/BRC-20, Lightning Network, Stacks L2), liquid staking, restaking, perpetuals, prediction markets (Polymarket, Kalshi), NFTs, RWAs, DAOs, cross-chain bridges, DePIN, and GameFi.
Before going full-time on-chain, I spent seven years in insurance doing financial investigations, multi-party reconciliation, and asset recovery. Recovered $3M+ running a nationwide subrogation portfolio and handled claims bordereaux across international counterparties in multiple currencies. I build what I need — Python, SQL, LLM APIs, whatever gets the job done. I'd rather get the work right than talk about it.
Most people in crypto can't do insurance. Most people in insurance can't do crypto. I do both, and I build the tools on top.
Independent on-chain research since 2017. Wallet clustering, fund-flow tracing, market manipulation detection, and protocol-level risk assessment across most major ecosystems. Deep hands-on participation across BTC (Ordinals, Lightning, Stacks), DeFi, NFTs, liquid staking, perpetuals, prediction markets, RWAs, DAOs, bridges, DePIN, and GameFi.
7+ years across commercial and personal lines. 7+ years handling claims end-to-end — subrogation recovery, reinsurance claims reporting, and multi-party reconciliation across 5+ countries including UAE, US, Qatar, Lebanon, and France.
Built working AI systems that solve real problems — document processing with LLM APIs (OpenAI, Anthropic, Gemini), RAG pipelines for cited answers, PII redaction workflows, and a video generation pipeline (Gemini + ElevenLabs + Shotstack). Python and LangChain are my primary tools.
Real numbers from real work. Insurance recovery data is from Crawford. On-chain coverage reflects actual activity since 2017.
Started in the UAE doing reinsurance, moved to the US for subrogation, went on-chain in 2017 and never looked back.
Started alongside employment in 2017. Went full-time in 2020 while handling family caregiving. Built everything independently — no team, no employer, no formal training. Grew $5K of personal capital to over $300K through on-chain analysis, position management, and risk controls I developed myself.
Ran the 7-Eleven subrogation portfolio nationwide. Owned every case from investigation through closure — all mine.
Everything here I built myself. No team, no employer, no tutorial.
Takes messy financial and legal PDFs, redacts PII on the browser side, chunks them through LLMs, and gives back structured answers with exact citations — page, paragraph, line number. Built it because I got tired of spending hours digging through documents manually.
Local AI running entirely on my machine — nothing leaves the device. Built the full environment from scratch. Use it for document analysis and research when I don't want data hitting external APIs.
Scrapes trending products, scores them by virality and commission potential, then auto-generates video reviews — Gemini writes the script, ElevenLabs does the voiceover, Shotstack renders the video. Currently deployed and running.
Custom SQL dashboards tracking on-chain activity: wallet clustering, DeFi flow analysis, token launch monitoring, and bridge activity surveillance. Original files lost due to hardware failure — methodology and SQL patterns retained.
Custom TradingView indicators built from my own trading strategies developed over years of live market testing. Momentum signals, on-chain divergence alerts, and liquidity zone detection. Not off-the-shelf TA — original methodology refined through thousands of trades and backtesting.
Custom multi-sheet Excel framework built to manage 300–400 concurrent subrogation files. Improved throughput by ~30%.
Cross-platform arbitrage and data-backed betting across Polymarket (2+ years) and Kalshi. Built Dune SQL dashboards with whale watchlists and dollar-threshold alerts to track big buys in real time. Custom Python scripts + AI pipeline scrapes and processes data faster than markets reprice. Covers crypto events, politics, sports, and economics — the edge comes from combining multiple signals, not any single one.
Python + LLM pipeline for building automated blockchain compliance reports — integrating Arkham wallet labels, risk scoring, and LLM-generated narrative summaries for flagged entities.
Open to on-chain intelligence, crypto compliance, InsurTech AI, and senior claims/subrogation roles. Available for remote or Dallas-based positions.