3x
faster chat replies on the support agent. A 45-second wait ran on every message. Now it's 15 seconds mid-conversation.
I make LLM agents work with real customers, and I measure it.
Most AI agents look great in a demo and fall apart with real customers. I'm the engineer who finds out why and fixes it, with numbers to prove the fix worked.
The search fix
13 of 23
test questions now find a document the agent used to miss. The right one ranks 1st or 2nd instead of 3rd.
The metric that lied
The dashboard number swung 4x in three weeks while the bot hadn't changed at all. I replaced it with a metric that counts real resolutions: no escalation, no reopen within 7 days, no human reply.
60→5
minutes for a credit analysis: about an hour before, as little as 5 with my pipeline over 5+ credit bureaus.
10→100
credit operations a day: about 10 before, up to 100 once the heavy work moved to RabbitMQ and Celery workers.
0.0205
MAE, the best result of my research, from plain DES on visceral leishmaniasis with HIV. No single model won everywhere.
Work
AI support agent for a B2B SaaS
It answers customers on WhatsApp, web chat and email. I shipped about 30 releases in 4 weeks, and replies got 3x faster.
Credit analysis pipeline
I rebuilt a credit fund's manual credit analysis as Python services with queues and workers.
LogStream
My first Go project. A sidecar agent streams logs over gRPC to a live dashboard, and healthmon blocks new requests to an external API once its error rate crosses a threshold.
RAG side projects
A PDF notes app and a resume generator on ChromaDB, plus Kai, my personal copilot, with embeddings and cosine similarity in numpy. The bug I remember: every restart wiped the vector index.
Tuberculosis surveillance research
Scrapers for Brazil's public health database feed a dashboard and a forecasting study. No single model won everywhere.
Write to me
anthonympodev@gmail.comI'm open to remote backend and AI engineering roles with US and European teams. I also take small freelance projects through my company, Insight AI & Automation.