hi, i'm Lakshmi Anne
principal ai engineer · ai systems architect
9+ years building production ai — from agentic llm platforms to real-time ml at enterprise scale.
open source work
flagship production-grade reference architecture, plus selected projects
enterprise-ai-platform
enterprise ai workflow platform
production reference architecture for multi-tenant rag with tool-using agents
- ▸multi-tenant saas with postgres row-level security
- ▸transactional outbox + kafka for async ingestion
- ▸pgvector hnsw semantic search with citation tracking
- ▸claude sonnet 4.6 agent with custom react loop (no langgraph)
- ▸sse-streamed chat with token, tool_use, citations events
- ▸full observability: opentelemetry, prometheus, loki, tempo, grafana
9 architecture decision records (adr) documenting trade-offs
rag-enterprise-search
→production-grade rag pipeline for enterprise document search
multi-cloud-ai-agent
→ai agent system that orchestrates llms across azure, aws, and gcp
mlops-azure-template
→end-to-end mlops template for ml model deployment on azure
llm-sql-to-python
→llm-powered tool that converts sql queries to python code
sentiment-kafka-pipeline
→real-time sentiment analysis pipeline using kafka and ml models
yolov8-object-tracking
→real-time object detection and tracking using yolov8 and deepsort
production ai at scale
ai gateway & agentic vulnerability reporting
agentic llm reporting platform for automated penetration testing. deterministic pipeline where rules own every fact — cves, cvss, severity — and llms only write prose. a hard gate structurally blocks hallucinated findings before they reach client reports.
real-time fraud detection
1m+ transactions/day on kafka with sub-100ms decisioning. xgboost + isolation forest ensemble.
gpt-4 hospitality platform
30+ uk hotels, 200k+ reviews analysed. rag over guest data for personalised responses.
demand forecasting
£2m annual savings across 10k+ skus. prophet + xgboost ensemble with weather signals.
storm risk modelling
£1.5m insurance claims avoided through early warning systems and route optimisation.
what i build
end-to-end ai systems — from the model layer to production, with governance built in
rag & semantic search
retrieval pipelines with pgvector / faiss, citation tracking, and grounded generation over enterprise knowledge.
agentic & multi-agent
tool-calling agents, bounded autonomy, and langgraph-style orchestration with deterministic control planes.
llm apps & structured outputs
schema-validated generation, structured extraction, and reliable llm calls on bedrock / claude / openai.
evaluation & guardrails
automated qa gates, fact-grounding against authoritative sources, and eval harnesses that catch regressions.
secure-by-design ai
owasp llm top 10, prompt-injection defence, trust boundaries, and data-residency controls on untrusted inputs.
mlops & production ml
real-time inference, streaming pipelines, containerised deployment, and observability at enterprise scale.
tools i use to ship
9+ years building production ai
i'm a principal ai engineer based in milton keynes, uk. i design and ship production ai across the stack — agentic and multi-agent llm systems, rag, structured outputs, and real-time ml — with guardrails, evaluation, and secure-by-design engineering built in from day one.
my work spans regulated industries (cybersecurity, banking, hospitality, retail, energy) where ai failures carry real consequences — from agentic llm systems and rag to real-time ml at enterprise scale.
let's build something
open to senior ai/ml roles, consulting engagements, and interesting collaborations. best way to reach me is email.