About Sam
AI Solutions Architect and hands-on lead engineer building production-grade GenAI workflows on Google Cloud (Cloud Functions Gen2, Cloud Tasks, Firestore). I design and ship agentic research/ETL systems with strong reliability patterns (rate-limit handling, retries/backoff, idempotency), cost governance (token/cost accounting, model routing), and auditable provenance (source traceability). Currently in pilot/testing phases with big 4 consultancies. Founder-operator background with experience taking systems from prototype to scalable, monitored production. I specialise in AI-augmented development, allowing me to architect and ship complex enterprise systems solo in < 6 months that typically require a 10+ person engineering team to deliver in 12.
English
Native or bilingual
Remote only
Primarily works remotely
Experience
- ecozeAIFounder, AI Solutions Architect & Lead EngineerTECHJanuary 2024 - Today (2 years and 5 months)Leeds, United KingdomBuilding the core platform for an automated carbon-data + supply chain mapping engine and evidence-backed research workflows for enterprise use cases.Architected and implemented serverless GenAI workflow orchestration on GCP (Cloud Functions Gen2 + Cloud Tasks + Firestore) to run long-running, multi-step research tasks reliably at scale.Built a unified multi-provider LLM runtime on Vertex (Gemini + OpenAI-compatible models) including OAuth-based auth, streaming response parsing, and shared token/cost accounting.Implemented a resilient Deep Research client with stream parsing, resumable reconnection, and post-completion polling to guarantee outputs and usage in long-running workflows.Engineered reliability under API volatility: robust retry policies to reduce failures and operator intervention.Built token and cost governance to track cost-per-transaction and apply model-routing policies (cheap-first then escalate) to maintain predictable unit economics.Developed an evidence pipeline: harvest citations from grounding metadata, unwrap redirect URLs, and generate per-URL usage annotations to support audit/assurance review.
- ecozeAI & University of Bristol (Faculty of Engineering)Technical Research Lead (JointTECHJune 2024 - August 2025 (1 year and 2 months)Bristol, UKSpearheaded a joint R&D initiative to standardise retail product data and improve classification quality.
- Defined technical roadmap and supervised MSc Data Science research to process and standardise 75,900+ heterogeneous product entries from major UK retailers.
- Oversaw a three-tier classification system (rule-based NLP, statistical models, LLM-assisted validation) to identify recyclable packaging materials, achieving 97% accuracy (as reported by the project).
- Translated research outputs into commercially viable, scalable GCP serverless workloads.
- ECOZE (B2C App)Founder & Senior Full Stack EngineerTECHJuly 2021 - December 2023 (2 years and 5 months)Leeds, United KingdomBuilt a consumer sustainability product and subsequently pivoted core IP and learnings into enterprise workflows.
- Built an automated travel tracking engine using smartphone telemetry (accelerometer/gyroscope/GPS) to infer transport mode (e.g., train vs car) with minimal user input.
- Implemented purchase tracking pipelines to ingest financial data and match transactions to product databases to estimate personal footprint.
Recommendations
Be the first to recommend Sam
Help this freelancer shine by sharing your experience working together.
These freelancer profiles also match your criteria
Agatha Frydrych
Backend Java Software Engineer
4.7
(3)
2
Baptiste Duhen
Fullstack developer
4.6
(4)
5
Amed Hamou
Senior Lead Developer
4
(2)
7
Audrey Champion
Web developer
4.3
(3)
4
Education
- BSc (Hons)University of BathBSc (Hons)