We are presenting the current progress on the titanium (Ti) hot-electron transition-edge devices. The ultimate goal of this work is to develop a submillimeter Hot-Electron Direct Detector (HEDD) with the noise equivalent power NEP = 10-18-10-20 W/Hz1/2 for the moderate resolution spectroscopy and Cosmic Microwave Background (CMB) studies on future space telescope (e.g., SPICA, SAFIR, SPECS, CMBPol) with cryogenically cooled (∼ 4-5 K) mirrors. Recentlyi, we have achieved the extremely low thermal conductance (∼ 20 fW/K at 300 mK and ∼ 0.1 fW/K at 40 mK) due to the electron-phonon decoupling in Ti nanodevices with niobium (Nb) Andreev contacts. This thermal conductance translates into the "phonon-noise" NEP ≈ 3×10-21 W/Hz1/2 at 40 mK and NEP ≈ 3×10-19 W/Hz1/2 at 300 mK. These record data indicate the great potential of the hot-electron detector for meeting many application needs. Beside the extremely low phonon-noise NEP, the nanobqlometers have a very low electron heat capacitance that makes them promising as detectors of single THz photonsii. As the next step towards the practical demonstration of the HEDD, we fabricated and tested somewhat larger than in Ref.1 devices (∼ 6 μm × 0.35 μm × 40 nm) whose critical temperature is well reproduced in the range 300-350 mK. The output electrical noise measured in these devices with a low-noise dc SQUID is dominated by the thermal energy fluctuations (ETF) aka "phonon noise". This indicates the high electrothermal loop gain that effectively suppresses the contributions of the Johnson noise and the amplifier (SQUID) noise. The electrical NEP = 6.7×10-18 W/Hz1/2 derived from these measurements is in good agreement with the predictions based on the thermal conductance data. The very low NEP and the high speed (∼ μs) are a unique combination not found in other detectors.