# Bill Software Engineer GenAI Systems Architect

> 🏗️Agentic AI Systems & Workflows  🎯𝗚𝗲𝗻𝗔𝗜 • 𝗥𝗔𝗚 • 𝗟𝗟𝗠  ⚡️Distributed Microservices • Data Pipeline • API & Infrastructure   🤖𝗣𝘆𝘁𝗵𝗼𝗻 • 𝗚𝗼 • 𝗔𝗪𝗦 • 𝗚𝗖𝗣

Location: San Francisco Bay Area, United States
Profile: https://flows.cv/billsoftwareengineergenaisyste

𝐈 𝐛𝐮𝐢𝐥𝐝 𝐀𝐈 𝐒𝐲𝐬𝐭𝐞𝐦𝐬 𝐟𝐨𝐫 𝐜𝐨𝐦𝐩𝐚𝐧𝐢𝐞𝐬 𝐚𝐧𝐝 𝐢𝐧𝐯𝐞𝐬𝐭𝐨𝐫𝐬 𝐰𝐡𝐨 𝐝𝐨𝐧'𝐭 𝐡𝐚𝐯𝐞 𝐭𝐢𝐦𝐞 𝐟𝐨𝐫 𝐡𝐲𝐩𝐞.
 
I am a 𝐏𝐫𝐨𝐟𝐞𝐬𝐬𝐢𝐨𝐧𝐚𝐥 𝐒𝐨𝐟𝐭𝐰𝐚𝐫𝐞 𝐄𝐧𝐠𝐢𝐧𝐞𝐞𝐫 and 𝐀𝐩𝐩𝐥𝐢𝐞𝐝 𝐀𝐈 𝐀𝐫𝐜𝐡𝐢𝐭𝐞𝐜𝐭 with 15+ years of shipping high-stakes production systems. I build the infrastructure that makes AI reliable at scale, not the models themselves. I live at the intersection of backend reliability, data engineering, and Large Language Models, designing 𝐀𝐠𝐞𝐧𝐭𝐢𝐜 𝐖𝐨𝐫𝐤𝐟𝐥𝐨𝐰𝐬 that solve complex operational bottlenecks: automated triage engines for SREs, decision-support tools for investors, and multi-source reasoning pipelines.
 
My niche is turning vague "we want GenAI" goals into 𝐃𝐞𝐭𝐞𝐫𝐦𝐢𝐧𝐢𝐬𝐭𝐢𝐜, 𝐎𝐛𝐬𝐞𝐫𝐯𝐚𝐛𝐥𝐞 𝐒𝐲𝐬𝐭𝐞𝐦𝐬. I focus on the "Day 2" problems that kill AI projects: Evaluation, Latency, Cost Optimization, and Production Guardrails.
 
👉 𝐁𝐮𝐢𝐥𝐝𝐞𝐫 / 𝐒𝐡𝐢𝐩𝐩𝐞𝐫: I take projects from blank doc to production. My stack is Python, Go, and TypeScript running on AWS/GCP. I care about SLOs, on-call sanity, and the total cost of inference.
 
👉 𝐒𝐜𝐚𝐥𝐞 & 𝐑𝐞𝐥𝐢𝐚𝐛𝐢𝐥𝐢𝐭𝐲: Hands-on with high-throughput APIs, event-driven architectures (Kafka), and 𝐄𝐓𝐋 𝐩𝐢𝐩𝐞𝐥𝐢𝐧𝐞𝐬 that keep stochastic models behaving reliably at scale.
 
👉 𝐏𝐫𝐨𝐝𝐮𝐜𝐭𝐢𝐨𝐧 𝐀𝐈: I build 𝐌𝐮𝐥𝐭𝐢-𝐋𝐚𝐲𝐞𝐫𝐞𝐝 𝐑𝐀𝐆 𝐏𝐢𝐩𝐞𝐥𝐢𝐧𝐞𝐬 (Vector + LLM Reranking), semantic caching, and custom eval suites. I combine pattern-matching with LLM-based PII scrubbing for end-to-end data privacy.
 
👉 𝐈𝐧𝐧𝐨𝐯𝐚𝐭𝐢𝐨𝐧: Primary inventor on a conversational AI patent and architect of CODiE-winning AI products.
 
𝐔𝐒 𝐖𝐨𝐫𝐤 𝐀𝐮𝐭𝐡: TN Visa (Canadian), W-2 Full-time. US-Remote or Bay Area Hybrid.
 
🤖 𝐂𝐨𝐦𝐩𝐨𝐮𝐧𝐝 𝐀𝐈: Multi-Agent Orchestration • MCP (Model Context Protocol) • Tool Use • Reasoning Chains • Structured Output 
🔎 𝐑𝐞𝐭𝐫𝐢𝐞𝐯𝐚𝐥: Hybrid Search • LLM Reranking • Vector Databases • Embeddings • Knowledge Graphs • Semantic Search 
⚡️ 𝐀𝐈 𝐄𝐧𝐠𝐢𝐧𝐞𝐞𝐫𝐢𝐧𝐠: MLOps • Evals • Inference Optimization • Prompt Engineering • Guardrails • LangChain 
☁️ 𝐁𝐚𝐜𝐤𝐞𝐧𝐝 & 𝐈𝐧𝐟𝐫𝐚: Kubernetes • Docker • Terraform • Redis • Elasticsearch • GraphQL • CI/CD • Serverless

## Work Experience
### ✔️ Staff Software Engineer (Stability & Performance) @ Spotnana
Jan 2025 – Present | San Francisco Bay Area
I designed and shipped Spotnana's 𝐀𝐈 𝐓𝐫𝐢𝐚𝐠𝐞 𝐒𝐲𝐬𝐭𝐞𝐦, a fully autonomous pipeline that monitors incoming production incidents, automatically triggers multi-layer root cause analysis, and delivers actionable diagnostics without human intervention, combining LLMs with real-time observability data across a distributed microservices backend.
🔹 𝗕𝘂𝗶𝗹𝘁 𝗮 𝗧𝘄𝗼-𝗦𝘁𝗮𝗴𝗲 𝗛𝘆𝗯𝗿𝗶𝗱 𝗥𝗔𝗚 𝗲𝗻𝗴𝗶𝗻𝗲 (Vector Retrieval + LLM Reranking) to cross-reference incoming incidents against thousands of historical resolutions for automated precedent matching.
🔹 𝗘𝗻𝗴𝗶𝗻𝗲𝗲𝗿𝗲𝗱 𝗺𝘂𝗹𝘁𝗶-𝘀𝗼𝘂𝗿𝗰𝗲 𝗱𝗮𝘁𝗮 𝗼𝗿𝗰𝗵𝗲𝘀𝘁𝗿𝗮𝘁𝗶𝗼𝗻 that dynamically fetches and synthesizes context from application logs, distributed traces, and raw supplier API payloads to surface silent failures that traditional monitoring misses.
🔹 𝗕𝘂𝗶𝗹𝘁 𝗮𝗻 𝗟𝗟𝗠-𝗱𝗿𝗶𝘃𝗲𝗻 𝗘𝗧𝗟 𝗽𝗶𝗽𝗲𝗹𝗶𝗻𝗲 to pre-process and structure noisy historical ticket data into high-signal embeddings for a specialized similarity search database.
🔹 𝗜𝗺𝗽𝗹𝗲𝗺𝗲𝗻𝘁𝗲𝗱 𝗹𝗮𝘆𝗲𝗿𝗲𝗱 𝗣𝗜𝗜 𝘀𝗰𝗿𝘂𝗯𝗯𝗶𝗻𝗴 (client-level filtering, local pattern-matching, and contextual LLM scrubbing) for SOC2-compliant handling of sensitive customer data.
🔹 𝗕𝘂𝗶𝗹𝘁 𝗮 𝗳𝘂𝗹𝗹-𝘀𝘁𝗮𝗰𝗸 𝗨𝗫 𝗹𝗮𝘆𝗲𝗿 𝗮𝗻𝗱 𝗦𝗹𝗮𝗰𝗸/𝗝𝗶𝗿𝗮 𝗶𝗻𝘁𝗲𝗴𝗿𝗮𝘁𝗶𝗼𝗻 providing real-time automated triaging, manual-trigger capabilities, and an interactive context-aware Q&A interface for deep-dive investigation.
🔹 𝗩𝗲𝗿𝗶𝗳𝗶𝗮𝗯𝗹𝗲 𝗶𝗺𝗽𝗮𝗰𝘁: 4x triage throughput at 2x resolution speed vs. manual baseline, with no quality degradation (comparable reopen rates). Resolved 88 production bugs in one month vs. 21 per engineer on average without using AI.
🛠️ 𝗧𝗲𝗰𝗵: Java • Python • Kafka • gRPC • Temporal • AWS • PostgreSQL • Cassandra • Kibana • Zipkin • Datadog • Jira API • Slack API • OpenAI API

### ✔️ Founding Engineer (AI/ML) @ RCY Data Mining
Jan 2024 – Present | San Francisco Bay Area
🔹 Build and run an AI/ML lab focused on decision-support tools for equity portfolios and pari-mutuel / event markets.
🔹 Built a deterministic EV/odds engine for live pari-mutuel / event markets (banded odds, shrinkage, minimum-N guards) that shows consistently positive ROI in both out-of-sample backtests and real-money betting.
🔹 Designed an AI-assisted research & risk pipeline for a concentrated equity portfolio — factor tables, dashboards, and sizing rules that have outperformed broad equity benchmarks on a risk-adjusted basis.
🔹 Use LLMs + retrieval (RAG) to turn raw market data into bilingual (EN/ZH) decision memos for the investor, tying “what changed” to real positions and portfolio actions.

### ✔️ Senior Software Engineer (API Platform) @ Samsara
Jan 2022 – Jan 2024 | San Francisco Bay Area
Senior Software Engineer on the API Platform team for Samsara’s Connected Operations Cloud, which processes trillions of data points and tens of billions of API calls every year across fleets, equipment, sites, and sensors.

🔹Owned core pieces of the public REST/GraphQL API stack: reverse proxy, auth, org-to-cell routing, and rate limiting for high-volume customers integrating TMS, ERP, and analytics tools.
🔹Designed and shipped features to make the API layer more reliable and observable: timeouts, scoped auth, leaky-bucket rate limits backed by Redis, better request logging, and dashboarding in Datadog / CloudWatch / Databricks. 
🔹Led the engineering work for a certified Microsoft Power BI custom connector: OAuth2 auth, schema design, pagination, and performance tuning so customers could self-serve Samsara telemetry into BI without custom ETL. 
🔹Worked across 300+ microservices and a cell-based, multi-region architecture, coordinating changes with service owners to keep APIs backwards-compatible while evolving schemas and adding new endpoints.
🔹Regularly on PagerDuty for 24/7 production support: led incident response for API outages and customer-visible regressions, did deep dives using logs + traces + data lake queries, and drove follow-up fixes into the platform.

### ✔️Principal Solutions Architect @ Automation Anywhere
Jan 2019 – Jan 2022 | San Francisco Bay Area
🔹 Work with business partners to create joint web application, SaaS, and data integration solutions.
🔹 Create new product ideas, write patent, and create project prototypes then see it all the way to market.  See section on patent for information on using Generative AI (ChatGPT) for RPA processes.
🔹 Hands-on 360° product development from inception to release using an assortment of technologies that includes: Node, JS, APEX, C#, Azure, AWS, API, RESTful API, Salesforce APEX and others.
🔹 Build corporate partnership and communicate effectively with non-technical business stakeholders both inside and outside the company.  Facilitate partnership discussion and make Power Point presentation for company Executives.
🔹 Organize development process and mentor junior team members, interview potential hires.  SCRUM master.  Team lead.

### ✔️Lead Software Engineer @ Conga
Jan 2017 – Jan 2018 | Silicon Valley, California
🔹 Lead a team of Engineers with hands-on development to create a world-first next-gen virtual assistant for business using advanced artificial intelligence (AI) and machine learning (ML) using Node and Azure.  
🔹 Integrated IBM Watson and Microsoft Luis to understand natural human speech at greater than 99% accuracy (NLU).
🔹 Created a seamless augmented reality (AR) using the latest machine learning technologies on mobile.  Allow customers to naturally converse with their business software while driving.
🔹 Mimic human speech with machine learning (ML) similar to Amazon Alexa.  Use Microsoft Bot Framework to orchestrate dynamic, multi-layered conversations.

### ✔️Senior Software Engineer @ Insightly
Jan 2015 – Jan 2017 | San Francisco, California
🔹 Write highly scalable and efficient web application that can grow with the company’s user base
🔹 Over 40k page-load per month, peak time of over 1000 hits per second
🔹 Use advanced algorithms to solve throughput problems that required database read of over 100k entries per job
🔹 Created an internal Assertion API that drive many different client-facing application and is called at least 10000 times a day

### ✔️Software Developer @ Ericsson
Jan 2014 – Jan 2015 | Mississauga, ON, Canada
🔹 Write software for extremely large scale solutions, SaaS based, data driven, marketed toward the Telecom Industry
🔹 Write code in Java, serve-side JavaScript.  Development in entire stack: database, logic, UI
🔹 Scrum master and team lead role, report to Project Manager.  Delegate tasks.  Go to meetings
🔹 Our software sells for at least $1million per client


## Education
### Bachelor of Science (B.S.) in Mathematics and Computer Science Double Major
Dalhousie University

### Machine Learning Engineering Camp in Computer Software Engineering
UC San Diego Extended Studies

### Certificate Program in Cloud Computing & Enterprise Data Analytics (Big Data)
University of Toronto

### High School Diplomas in General Education
Queen Elizabeth Highschool


## Contact & Social
- LinkedIn: https://linkedin.com/in/billsoftwareengineer

---
Source: https://flows.cv/billsoftwareengineergenaisyste
JSON Resume: https://flows.cv/billsoftwareengineergenaisyste/resume.json
Last updated: 2026-04-01