Agentic Workflow Development

Independent Product R&D

Designing Secure Agent Transaction Infrastructure

A protocol-neutral reference design for giving AI agents bounded authority, exact approvals, isolated credentials, safe outcome handling and auditable transaction evidence.

Design outputs

The counts below describe elements in this target reference design and its public documentation; they are not observed production results or performance measurements.

hypREspace

hypREspace: Cutting Token Costs 90% While Raising Tool-Calling Accuracy to 95%

Dynamic prompt injection, a gold evaluation set and Amazon Nova fine-tuning cut a live agent's token spend by 90% and lifted tool-calling accuracy from 75.8% to 95%.

Measured outcomes

Measured with before-and-after per-query token counts and tool-call accuracy scored against the gold evaluation set.

hypREspace

hypREspace: Rearchitecting NL-to-SQL into an Agentic Analytics Engine

A brittle multi-hop prompt chain became one observable LangGraph ReAct agent with a custom MCP layer over the existing FastAPI backend and Claude on Bedrock.

Production outcomes

Production status reflects the deployed client workflow and its live traceability; no comparative performance figure is claimed.

Compliance-focused accounting technology startup (under NDA)

AI Research Platform for a Regulated Accounting Profession

End-to-end MVP architecture for cited, verified tax research answers: LangGraph agent, hybrid RAG, a claim verification gate and region-locked multi-tenant infrastructure.

Design targets

The 95% accuracy target, P50/P95 latency objectives, seven-day freshness target and eight-skill count are design objectives and reference-concept counts, not observed results.

Premium family connection platform (under NDA)

Agentic Question Generation for a Family Connection App

A LangGraph agent with agentic RAG generates personalised daily storytelling prompts as strict JSON, served to a live mobile app through FastAPI on AWS.

Production outcomes

Production status reflects the live mobile integration and operational tracing; no numerical engagement lift is claimed.

Secure Agent Transaction Infrastructure

Bounded authority for agents that can transact

Tool access does not by itself establish transaction authority. This protocol-neutral target reference architecture turns agent proposals into bounded, policy-checked, transaction-bound, credential-isolated and auditable external actions.

Adaptive RAG

Toggle-Driven Hybrid Retrieval With Skills

A self-hosted LangGraph and FastAPI RAG service for a regulated finance and legal practice. Vector search always runs; allowlisted web search is added only when the request toggles it on, never by an LLM router. Four self-reflection gates with hard-capped retry loops decide whether an answer ships. Slash-invocable skills run as isolated sub-agents over the same graph, and opt-in Langfuse tracing plus async DeepEval scoring observe every run without adding latency.

Fine-Tuned Tool-Calling Agent

A Small Model Taught to Use 40+ Tools

A LangGraph ReAct agent over a custom MCP backend exposing 40+ tools, with the decision model replaced by a fine-tuned Gemma 3 4B. Successful multi-turn tool trajectories from production were filtered and reformatted into an SFT dataset, then used to train Gemma 3 4B with QLoRA — cutting cost and latency while making tool-call formatting far more reliable.

LangChain DeepAgent

Production-Ready AI Agent with Specialized Skills

A FastAPI service powering a LangChain DeepAgent backed by Ollama llama3.2:3b. Features four specialized skills for complex reasoning tasks, SSE streaming for real-time responses, Redis rate limiting, and multi-turn session memory.