alex → cv

• Lead AI engineering for the analytics product, cutting analysis time by about 80% across the BI analyst team
• Designed and built the product's agent harness: 6 subagents and 50 tools, with reusable and scheduled workflows
• Built 1,000+ eval cases across subagents, tool use, end-to-end task completion, and guardrails
• Designed the agents' memory architecture (episodic, semantic, and procedural), unifying the semantic layer and workflows
• Built the agent that turns requests into scheduled workflows producing the weekly and monthly reviews sent to clients
• Built an internal portal for monitoring, debugging, and managing agent runs
• Extended the semantic data model with ontological abstractions and business concepts that root-cause analysis and data discovery are built on, and prototyped the internal tool for managing it
• Built the first insight agents on Jinja templates, with lineage and provenance tracing every generated number to its source
• Ran a fault-tolerant distributed worker fleet on AWS (ECS/SNS/SQS) with dead-letter queue handling and zero-downtime deploys
• Built a suite of agentic coding plugins for the engineering team

• Owned AI and backend engineering, taking the product from prototype to production in 2 months
• Built hybrid semantic and graph retrieval to ground personalized, on-brand content in each customer's data
• Built the platform the AI ran on (API, data model, authentication), and onboarded an engineer onto DevOps and AWS
• Added monitoring across the AI pipelines to catch quality regressions
• Built ontological models to give the data model a consistent vocabulary
• Coordinated with an external app development team on integrating their app with the platform

• Built a knowledge-graph AI copilot over thousands of documents for a consulting firm: a graph UI, chatbot, and Claude Code plugin that query it live and write reports and decks
• Advise startups and established companies on AI strategy: technology assessments, build-versus-buy, and stack choices
• Gave a keynote to about 30 CIOs and founders on LLM architectures, guardrails, and feedback loops
• Codified organizational process for a blockchain unicorn: RACI matrices, documented processes, and access controls aligned across the org and built into internal tooling
• Built knowledge-management agents in Claude Code and Obsidian for founders, consultants, and creatives

• Led a cross-functional team of 5 to 10 engineers with the music, research, and sales teams
• Planned the roadmap and releases with the founders around customer and partner needs
• Built the pipelines that ingested labels' and publishers' catalogues for AI tagging and search
• Partnered with the music research team and in-house experts to design the labeling and model validation processes
• Ran GCP on Kubernetes and Istio with Grafana and Prometheus, and rebuilt CI/CD on Jenkins with Cypress tests

• Set technical direction and prototyped a supercomputer-emulation platform with my co-founder, launched in 14 months
• Raised a seed round from VCs and angels and grew the team to 15 across two offices
• Sold to national supercomputing centres and enterprise R&D teams across Asia-Pacific
• Set the business model, go-to-market strategy, and financial model
• Managed relationships with several levels of government across the region
• Ran investor relations and board communications
• Negotiated the contracts with partners and suppliers

• Co-founded Archanan out of the programme
• Secured letters of intent from early customers before committing to a build
• Tested several ideas with potential customers and dropped the ones that did not hold up
• Built the financial model and go-to-market plan
• Led the first fundraise
• Formed the company's early partnerships

• Built job scheduling optimization for grid-style workflows on commodity hardware, then ported it to cloud and bare metal
• Led engineering as the team grew from 2 to 20+ people, and set its code review and QA standards
• Directed the scheduling algorithms, and worked directly with clients on what to build next
• Raised seed capital from VCs and angels, and got the company into the Grand Central Tech accelerator

The Center of Excellence in Wireless Information Technology.

• Designed, prototyped, and tested audio synthesizers, including the circuit design
• Led seminars on music and mathematics
• Turned research into pieces that could be performed
• Collaborated on synthesis with Pigtronix, my startup Motiff Tech, and Stony Brook's digital arts center

• Researched optimal wavelet bases for audio compression in C++ on HPC clusters, running experiments that scored timbre coherence against compression ratio
• Ran a supercomputing project funded by the High Performance Computing Consortium of New York
• Maintained project documentation and datasets for other researchers

My first software job, and my first hardware job.

• Wrote and tested production firmware for digital guitar pedals, building the test fixtures
• Developed a digital tape loop emulator in Objective-C
• Turned customer service reports into technical reports, and worked with the engineering team to diagnose and triage the underlying issues
• Shipped firmware releases through CI and pre-release testing that caught defects before they reached customers
• Wrote the technical documentation and user manuals
• Fed customer feedback into product planning

• Assembled and tested circuit boards for audio processing units, and trained new staff on assembly and quality control
• Handled customer service, repairs, and returns as the shop's first point of contact