Mohsen Jahanshahi
Production AI and
automation systems.
I build and operate software products across AI infrastructure, market analytics, DeFi execution, developer APIs, and growth intelligence. The portfolio below shows breadth, shipped systems, and production discipline.
Engineering systems for reliable automation.
I design and ship production software across AI infrastructure, market analytics, DeFi execution, workflow automation, and growth intelligence. The focus is consistent delivery, measurable performance, and maintainable architecture.
Core implementations include semantic memory services, multi-channel agent runtimes, retrieval pipelines for technical document sets, strategy engines for market monitoring, and social-channel analytics pipelines for lead generation teams. Most systems are deployed as containerized services with clear observability and recovery paths.
Primary stack: TypeScript and Python, with Rust and Go for performance-sensitive workloads. Infrastructure is built for uptime, cost control, and data ownership.
Deployed Services
AI, analytics, and workflow systems
Commercial Verticals
DeFi, B2B, SaaS, growth, data
Active Repositories
Production and R&D tracks
Deployment snapshot
memory-servicehealthy
inference-gatewayhealthy
chat-bridgehealthy
api-proxyhealthy
event-workerhealthy
monitoringhealthy
Selected Work
Production systems and active builds
hari-hive (AgentSSOT)
Semantic Memory for AI Agents
Persistent knowledge graph that gives any AI agent long-term memory. Stores facts, skills, and decisions as embeddings in ChromaDB with scoped namespaces per agent. Every agent I run writes to and reads from this single source of truth. Cross-session context, skill teaching, and feedback loops that improve retrieval ranking over time. Saves 50%+ on token usage by eliminating repeated context.
Polymarket Profit Machine
Delta-Neutral Arbitrage on Solana
Five arbitrage detection strategies scanning prediction markets for mispriced contracts. Opens hedged positions through Jupiter perpetual contracts to capture funding rate differentials while staying delta-neutral. Configurable thresholds, automated Telegram alerts, and a dashboard for tracking P&L across strategies.
VectorData.solutions
Agentic AI for Engineering Firms
B2B platform ingesting engineering specs, BOMs, and datasheets into a RAG pipeline with agentic parts-lookup. High-accuracy sourcing across ODB device catalogs. Team project with active clients.
BrandForge
Growth Analytics and Lead Ops Automation
Multi-channel marketing operations agent for LinkedIn, Instagram, TikTok, and Farcaster workflows. Supports campaign planning, lead routing, content testing, and performance analytics.
Clawdbot Trend Engine
Social Signal Discovery for Commerce Teams
Continuous trend-detection pipeline scanning short-form social channels for high-velocity topics and product signals. Designed to surface early demand indicators for brand and commerce decisions.
CrossSite
Internal Network Sync Service
Lightweight synchronization service for internal networks. Text and image payloads sync across devices through web UI, CLI API, and real-time WebSocket transport. Built in Go.
Hairy
Multi-Channel Agent Harness
Standalone agent runtime with Telegram, WhatsApp, webhooks, and CLI channels. Self-growing skill registry with lifecycle promotion (draft to active to retired). JSON-RPC sidecars in Rust and Go for compute-heavy tasks.
ClawBox
Open-Source Agent Framework
Plugin-based framework for building AI agents. Multi-model routing, tool orchestration, and execution loops with self-improvement hooks. Each agent is a composition of swappable capabilities.
termwhat
AI Terminal Command Assistant
CLI utility that generates terminal commands from natural language prompts using private Ollama inference. Built for operator speed and reduced context switching.
VEIL Intelligence
Market Insights and Signal Platform
Analytics platform for market monitoring, signal scoring, and decision support across equities and digital assets. Includes data ingestion, alerting workflows, and execution-facing insight summaries.
SlopeSniper
Algorithmic Trading System
Technical analysis engine with automated backtesting, real-time signal generation, and position sizing. Strict code quality: pytest, ruff, mypy enforced on every commit.
prompt-forge
Idea-to-App Pipeline
Repeatable pipeline for transforming ideas into production apps with Claude Code. Structured prompts, scaffolding templates, and deployment automation.
Idea Park
Project Incubation SaaS
Platform for managing the pipeline from concept to shipped product. Stripe payment integration, user auth, structured stage gates. Tracks and prioritizes across 40+ concurrent projects.
Commercial API Access
Two APIs are currently in development and will launch in phases. Early access is invitation-only while reliability testing is in progress.
Northstar News API
PlannedGlobal news ingestion pipeline for structured headlines, source normalization, and relevance filtering. Built for teams that need clean news signals in product or dashboard workflows.
Sentinel Market Signals API
PlannedMarket insights feed covering watchlist movement, trend summaries, and signal metadata. Designed for analytics and decision-support use cases without exposing internal strategy logic.
Planned Pricing
Starter
- •Planned: 10,000 requests/day
- •News API access
- •Email support
Pro
- •Planned: 100,000 requests/day
- •News + Market Signals
- •Priority support
- •Webhook delivery
Enterprise
- •Dedicated limits
- •Custom SLA
- •Private onboarding
- •Contract support
What I Work With
Languages
Frontend
Backend
AI / ML
Blockchain
Infrastructure
DevOps
Growth Ops
Platforms
Infrastructure Practice
Environment design focused on uptime, observability, change control, and efficient compute usage.
Virtualization cluster with dedicated VMs per workload, service isolation, GPU passthrough, and snapshot-based rollback.
NVIDIA-backed inference tier using Ollama and vLLM. Model serving and embedding generation run on private infrastructure.
Cloudflare Tunnels for zero-trust ingress. NGINX reverse proxies per service. VLAN separation between production and experimental workloads.
Prometheus metrics collection, Grafana dashboards, and GPU utilization tracking via nvidia_gpu_exporter. Node exporter on every VM.
Automated backup strategy with daily database snapshots and recovery testing. PostgreSQL with pgvector for transactional and vector workloads.
Get in touch.
Available for consulting and build partnerships on AI infrastructure, API products, workflow automation, and systems architecture. If you need a technical operator who can design and ship end to end, reach out.
[email protected]© 2026 Mohsen Jahanshahi
Self-hosted on Docker · Cloudflare Tunnel · Next.js · FastAPI